Awesome Learning to Hash

The 2023 Big ANN Challenge, held at NeurIPS 2023, focused on advancing the state-of-the-art in indexing data structures and search algorithms for practical variants of Approximate Nearest Neighbor (ANN) search that reflect the growing complexity and diversity of workloads. Unlike prior challenges that emphasized scaling up classical ANN search ~\cite{DBLP:conf/nips/SimhadriWADBBCH21}, this competition addressed filtered search, out-of-distribution data, sparse and streaming variants of ANNS. Participants developed and submitted innovative solutions that were evaluated on new standard datasets with constrained computational resources. The results showcased significant improvements in search accuracy and efficiency over industry-standard baselines, with notable contributions from both academic and industrial teams. This paper summarizes the competition tracks, datasets, evaluation metrics, and the innovative approaches of the top-performing submissions, providing insights into the current advancements and future directions in the field of approximate nearest neighbor search.

We present the first provably sublinear time hashing algorithm for approximate Maximum Inner Product Search (MIPS). Searching with (un-normalized) inner product as the underlying similarity measure is a known difficult problem and finding hashing schemes for MIPS was considered hard. While the existing Locality Sensitive Hashing (LSH) framework is insufficient for solving MIPS, in this paper we extend the LSH framework to allow asymmetric hashing schemes. Our proposal is based on a key observation that the problem of finding maximum inner products, after independent asymmetric transformations, can be converted into the problem of approximate near neighbor search in classical settings. This key observation makes efficient sublinear hashing scheme for MIPS possible. Under the extended asymmetric LSH (ALSH) framework, this paper provides an example of explicit construction of provably fast hashing scheme for MIPS. Our proposed algorithm is simple and easy to implement. The proposed hashing scheme leads to significant computational savings over the two popular conventional LSH schemes: (i) Sign Random Projection (SRP) and (ii) hashing based on p-stable distributions for L2 norm (L2LSH), in the collaborative filtering task of item recommendations on Netflix and Movielens (10M) datasets.

The query complexity of locality sensitive hashing (LSH) based similarity search is dominated by the number of hash evaluations, and this number grows with the data size (Indyk & Motwani, 1998). In industrial applications such as search where the data are often high-dimensional and binary (e.g., text n-grams), minwise hashing is widely adopted, which requires applying a large number of permutations on the data. This is costly in computation and energy-consumption. In this paper, we propose a hashing technique which generates all the necessary hash evaluations needed for similarity search, using one single permutation. The heart of the proposed hash function is a “rotation” scheme which densifies the sparse sketches of one permutation hashing (Li et al., 2012) in an unbiased fashion thereby maintaining the LSH property. This makes the obtained sketches suitable for hash table construction. This idea of rotation presented in this paper could be of independent interest for densifying other types of sparse sketches. Using our proposed hashing method, the query time of a (K, L)-parameterized LSH is reduced from the typical O(dKL) complexity to merely O(KL + dL), where d is the number of nonzeros of the data vector, K is the number of hashes in each hash table, and L is the number of hash tables. Our experimental evaluation on real data confirms that the proposed scheme significantly reduces the query processing time over minwise hashing without loss in retrieval accuracies.

Cayley hash functions are based on a simple idea of using a pair of semigroup elements, A and B, to hash the 0 and 1 bit, respectively, and then to hash an arbitrary bit string in the natural way, by using multiplication of elements in the semigroup. The main advantage of Cayley hash functions compared to, say, hash functions in the SHA family is that when an already hashed document is amended, one does not have to hash the whole amended document all over again, but rather hash just the amended part and then multiply the result by the hash of the original document. Some authors argued that this may be a security hazard, specifically that this property may facilitate finding a second preimage by splitting a long bit string into shorter pieces. In this paper, we offer a way to get rid of this alleged disadvantage and keep the advantages at the same time. We call this method ``Cayley hashing with cookies” using terminology borrowed from the theory of random walks in a random environment. For the platform semigroup, we use 2x2 matrices over F_p.

Machine learning on source code (MLOnCode) is a popular research field that has been driven by the availability of large-scale code repositories and the development of powerful probabilistic and deep learning models for mining source code. Code-to-code recommendation is a task in MLOnCode that aims to recommend relevant, diverse and concise code snippets that usefully extend the code currently being written by a developer in their development environment (IDE). Code-to-code recommendation engines hold the promise of increasing developer productivity by reducing context switching from the IDE and increasing code-reuse. Existing code-to-code recommendation engines do not scale gracefully to large codebases, exhibiting a linear growth in query time as the code repository increases in size. In addition, existing code-to-code recommendation engines fail to account for the global statistics of code repositories in the ranking function, such as the distribution of code snippet lengths, leading to sub-optimal retrieval results. We address both of these weaknesses with Senatus, a new code-to-code recommendation engine. At the core of Senatus is De-Skew LSH a new locality sensitive hashing (LSH) algorithm that indexes the data for fast (sub-linear time) retrieval while also counteracting the skewness in the snippet length distribution using novel abstract syntax tree-based feature scoring and selection algorithms. We evaluate Senatus via automatic evaluation and with an expert developer user study and find the recommendations to be of higher quality than competing baselines, while achieving faster search. For example, on the CodeSearchNet dataset we show that Senatus improves performance by 6.7% F1 and query time 16x is faster compared to Facebook Aroma on the task of code-to-code recommendation.

Hashing is widely used to solve fast Approximate Nearest Neighbor (ANN) search problems, involves converting the original real-valued samples to binary-valued representations. The conventional quantization strategies, such as Single-Bit Quantization and Multi-Bit quantization, are considered ineffective, because of their serious information loss. To address this issue, we propose a novel variable-length quantization (VLQ) strategy for hashing. In the proposed VLQ technique, we divide all samples into different regions in each dimension firstly given the real-valued features of samples. Then we compute the dispersion degrees of these regions. Subsequently, we attempt to optimally assign different number of bits to each dimensions to obtain the minimum dispersion degree. Our experiments show that the VLQ strategy achieves not only superior performance over the state-of-the-art methods, but also has a faster retrieval speed on public datasets.

Recent vision and learning studies show that learning compact hash codes can facilitate massive data processing with significantly reduced storage and computation. Particularly, learning deep hash functions has greatly improved the retrieval performance, typically under the semantic supervision. In contrast, current unsupervised deep hashing algorithms can hardly achieve satisfactory performance due to either the relaxed optimization or absence of similarity-sensitive objective. In this work, we propose a simple yet effective unsupervised hashing framework, named Similarity-Adaptive Deep Hashing (SADH), which alternatingly proceeds over three training modules: deep hash model training, similarity graph updating and binary code optimization. The key difference from the widely-used two-step hashing method is that the output representations of the learned deep model help update the similarity graph matrix, which is then used to improve the subsequent code optimization. In addition, for producing high-quality binary codes, we devise an effective discrete optimization algorithm which can directly handle the binary constraints with a general hashing loss. Extensive experiments validate the efficacy of SADH, which consistently outperforms the state-of-the-arts by large gaps.

Semantic hashing has become a powerful paradigm for fast similarity search in many information retrieval systems. While fairly successful, previous techniques generally require two-stage training, and the binary constraints are handled ad-hoc. In this paper, we present an end-to-end Neural Architecture for Semantic Hashing (NASH), where the binary hashing codes are treated as Bernoulli latent variables. A neural variational inference framework is proposed for training, where gradients are directly backpropagated through the discrete latent variable to optimize the hash function. We also draw connections between proposed method and rate-distortion theory, which provides a theoretical foundation for the effectiveness of the proposed framework. Experimental results on three public datasets demonstrate that our method significantly outperforms several state-of-the-art models on both unsupervised and supervised scenarios.

Conventional unsupervised hashing methods usually take advantage of similarity graphs, which are either pre-computed in the high-dimensional space or obtained from random anchor points. On the one hand, existing methods uncouple the procedures of hash function learning and graph construction. On the other hand, graphs empirically built upon original data could introduce biased prior knowledge of data relevance, leading to sub-optimal retrieval performance. In this paper, we tackle the above problems by proposing an efficient and adaptive code-driven graph, which is updated by decoding in the context of an auto-encoder. Specifically, we introduce into our framework twin bottlenecks (i.e., latent variables) that exchange crucial information collaboratively. One bottleneck (i.e., binary codes) conveys the high-level intrinsic data structure captured by the code-driven graph to the other (i.e., continuous variables for low-level detail information), which in turn propagates the updated network feedback for the encoder to learn more discriminative binary codes. The auto-encoding learning objective literally rewards the code-driven graph to learn an optimal encoder. Moreover, the proposed model can be simply optimized by gradient descent without violating the binary constraints. Experiments on benchmarked datasets clearly show the superiority of our framework over the state-of-the-art hashing methods.

Recent binary representation learning models usually require sophisticated binary optimization, similarity measure or even generative models as auxiliaries. However, one may wonder whether these non-trivial components are needed to formulate practical and effective hashing models. In this paper, we answer the above question by proposing an embarrassingly simple approach to binary representation learning. With a simple classification objective, our model only incorporates two additional fully-connected layers onto the top of an arbitrary backbone network, whilst complying with the binary constraints during training. The proposed model lower-bounds the Information Bottleneck (IB) between data samples and their semantics, and can be related to many recent `learning to hash’ paradigms. We show that, when properly designed, even such a simple network can generate effective binary codes, by fully exploring data semantics without any held-out alternating updating steps or auxiliary models. Experiments are conducted on conventional large-scale benchmarks, i.e., CIFAR-10, NUS-WIDE, and ImageNet, where the proposed simple model outperforms the state-of-the-art methods.

Contrastive Language and Image Pairing (CLIP), a transformative method in multimedia retrieval, typically trains two neural networks concurrently to generate joint embeddings for text and image pairs. However, when applied directly, these models often struggle to differentiate between visually distinct images that have similar captions, resulting in suboptimal performance for image-based similarity searches. This paper addresses the challenge of optimizing CLIP models for various image-based similarity search scenarios, while maintaining their effectiveness in text-based search tasks such as text-to-image retrieval and zero-shot classification. We propose and evaluate two novel methods aimed at refining the retrieval capabilities of CLIP without compromising the alignment between text and image embeddings. The first method involves a sequential fine-tuning process: initially optimizing the image encoder for more precise image retrieval and subsequently realigning the text encoder to these optimized image embeddings. The second approach integrates pseudo-captions during the retrieval-optimization phase to foster direct alignment within the embedding space. Through comprehensive experiments, we demonstrate that these methods enhance CLIP’s performance on various benchmarks, including image retrieval, k-NN classification, and zero-shot text-based classification, while maintaining robustness in text-to-image retrieval. Our optimized models permit maintaining a single embedding per image, significantly simplifying the infrastructure needed for large-scale multi-modal similarity search systems.

We show how to learn a deep graphical model of the word-count vectors obtained from a large set of documents. The values of the latent variables in the deepest layer are easy to infer and give a much better representation of each document than Latent Semantic Analysis. When the deepest layer is forced to use a small number of binary variables (e.g. 32), the graphical model performs “semantic hashing”: Documents are mapped to memory addresses in such a way that semantically similar documents are located at nearby addresses. Documents similar to a query document can then be found by simply accessing all the addresses that differ by only a few bits from the address of the query document. This way of extending the efficiency of hash-coding to approximate matching is much faster than locality sensitive hashing, which is the fastest current method. By using semantic hashing to filter the documents given to TF-IDF, we achieve higher accuracy than applying TF-IDF to the entire document set.

Adversarial examples add imperceptible alterations to inputs with the objective to induce misclassification in machine learning models. They have been demonstrated to pose significant challenges in domains like image classification, with results showing that an adversarially perturbed image to evade detection against one classifier is most likely transferable to other classifiers. Adversarial examples have also been studied in malware analysis. Unlike images, program binaries cannot be arbitrarily perturbed without rendering them non-functional. Due to the difficulty of crafting adversarial program binaries, there is no consensus on the transferability of adversarially perturbed programs to different detectors. In this work, we explore the robustness of malware detectors against adversarially perturbed malware. We investigate the transferability of adversarial attacks developed against one detector, against other machine learning-based malware detectors, and code similarity techniques, specifically, locality sensitive hashing-based detectors. Our analysis reveals that adversarial program binaries crafted for one detector are generally less effective against others. We also evaluate an ensemble of detectors and show that they can potentially mitigate the impact of adversarial program binaries. Finally, we demonstrate that substantial program changes made to evade detection may result in the transformation technique being identified, implying that the adversary must make minimal changes to the program binary.

The fruit fly Drosophila’s olfactory circuit has inspired a new locality sensitive hashing (LSH) algorithm, FlyHash. In contrast with classical LSH algorithms that produce low dimensional hash codes, FlyHash produces sparse high-dimensional hash codes and has also been shown to have superior empirical performance compared to classical LSH algorithms in similarity search. However, FlyHash uses random projections and cannot learn from data. Building on inspiration from FlyHash and the ubiquity of sparse expansive representations in neurobiology, our work proposes a novel hashing algorithm BioHash that produces sparse high dimensional hash codes in a data-driven manner. We show that BioHash outperforms previously published benchmarks for various hashing methods. Since our learning algorithm is based on a local and biologically plausible synaptic plasticity rule, our work provides evidence for the proposal that LSH might be a computational reason for the abundance of sparse expansive motifs in a variety of biological systems. We also propose a convolutional variant BioConvHash that further improves performance. From the perspective of computer science, BioHash and BioConvHash are fast, scalable and yield compressed binary representations that are useful for similarity search.

We seek to build a large collection of images with ground truth labels to be used for object detection and recognition research. Such data is useful for supervised learning and quantitative evaluation. To achieve this, we developed a web-based tool that allows easy image annotation and instant sharing of such annotations. Using this annotation tool, we have collected a large dataset that spans many object categories, often containing multiple instances over a wide variety of images. We quantify the contents of the dataset and compare against existing state of the art datasets used for object recognition and detection. Also, we show how to extend the dataset to automatically enhance object labels with WordNet, discover object parts, recover a depth ordering of objects in a scene, and increase the number of labels using minimal user supervision and images from the web.

This extended abstract explores the integration of federated learning with deep transfer hashing for distributed prediction tasks, emphasizing resource-efficient client training from evolving data streams. Federated learning allows multiple clients to collaboratively train a shared model while maintaining data privacy - by incorporating deep transfer hashing, high-dimensional data can be converted into compact hash codes, reducing data transmission size and network loads. The proposed framework utilizes transfer learning, pre-training deep neural networks on a central server, and fine-tuning on clients to enhance model accuracy and adaptability. A selective hash code sharing mechanism using a privacy-preserving global memory bank further supports client fine-tuning. This approach addresses challenges in previous research by improving computational efficiency and scalability. Practical applications include Car2X event predictions, where a shared model is collectively trained to recognize traffic patterns, aiding in tasks such as traffic density assessment and accident detection. The research aims to develop a robust framework that combines federated learning, deep transfer hashing and transfer learning for efficient and secure downstream task execution.

In this work, we report on a novel application of Locality Sensitive Hashing (LSH) to seismic data at scale. Based on the high waveform similarity between reoccurring earthquakes, our application identifies potential earthquakes by searching for similar time series segments via LSH. However, a straightforward implementation of this LSH-enabled application has difficulty scaling beyond 3 months of continuous time series data measured at a single seismic station. As a case study of a data-driven science workflow, we illustrate how domain knowledge can be incorporated into the workload to improve both the efficiency and result quality. We describe several end-toend optimizations of the analysis pipeline from pre-processing to post-processing, which allow the application to scale to time series data measured at multiple seismic stations. Our optimizations enable an over 100× speedup in the end-to-end analysis pipeline. This improved scalability enabled seismologists to perform seismic analysis on more than ten years of continuous time series data from over ten seismic stations, and has directly enabled the discovery of 597 new earthquakes near the Diablo Canyon nuclear power plant in California and 6123 new earthquakes in New Zealand.

We propose two efficient approximations to standard convolutional neural networks: Binary-Weight-Networks and XNOR-Networks. In Binary-Weight-Networks, the filters are approximated with binary values resulting in 32x memory saving. In XNOR-Networks, both the filters and the input to convolutional layers are binary. XNOR-Networks approximate convolutions using primarily binary operations. This results in 58x faster convolutional operations and 32x memory savings. XNOR-Nets offer the possibility of running state-of-the-art networks on CPUs (rather than GPUs) in real-time. Our binary networks are simple, accurate, efficient, and work on challenging visual tasks. We evaluate our approach on the ImageNet classification task. The classification accuracy with a Binary-Weight-Network version of AlexNet is only 2.9\% less than the full-precision AlexNet (in top-1 measure). We compare our method with recent network binarization methods, BinaryConnect and BinaryNets, and outperform these methods by large margins on ImageNet, more than 16\% in top-1 accuracy.

In embedding-based retrieval, Approximate Nearest Neighbor (ANN) search enables efficient retrieval of similar items from large-scale datasets. While maximizing recall of relevant items is usually the goal of retrieval systems, a low precision may lead to a poor search experience. Unlike lexical retrieval, which inherently limits the size of the retrieved set through keyword matching, dense retrieval via ANN search has no natural cutoff. Moreover, the cosine similarity scores of embedding vectors are often optimized via contrastive or ranking losses, which make them difficult to interpret. Consequently, relying on top-K or cosine-similarity cutoff is often insufficient to filter out irrelevant results effectively. This issue is prominent in product search, where the number of relevant products is often small. This paper introduces a novel relevance filtering component (called “Cosine Adapter”) for embedding-based retrieval to address this challenge. Our approach maps raw cosine similarity scores to interpretable scores using a query-dependent mapping function. We then apply a global threshold on the mapped scores to filter out irrelevant results. We are able to significantly increase the precision of the retrieved set, at the expense of a small loss of recall. The effectiveness of our approach is demonstrated through experiments on both public MS MARCO dataset and internal Walmart product search data. Furthermore, online A/B testing on the Walmart site validates the practical value of our approach in real-world e-commerce settings.

As datasets grow larger, matching and merging entries from different databases has become a costly task in modern data pipelines. To avoid expensive comparisons between entries, blocking similar items is a popular preprocessing step. In this paper, we introduce BlockBoost, a novel boosting-based method that generates compact binary hash codes for database entries, through which blocking can be performed efficiently. The algorithm is fast and scalable, resulting in computational costs that are orders of magnitude lower than current benchmarks. Unlike existing alternatives, BlockBoost comes with associated feature importance measures for interpretability, and possesses strong theoretical guarantees, including lower bounds on critical performance metrics like recall and reduction ratio. Finally, we show that BlockBoost delivers great empirical results, outperforming state-of-the-art blocking benchmarks in terms of both performance metrics and computational cost.

This paper addresses the problem of designing binary codes for high-dimensional data such that vectors that are similar in the original space map to similar binary strings. We introduce a simple distribution-free encoding scheme based on random projections, such that the expected Hamming distance between the binary codes of two vectors is related to the value of a shift-invariant kernel (e.g., a Gaussian kernel) between the vectors. We present a full theoretical analysis of the convergence properties of the proposed scheme, and report favorable experimental performance as compared to a recent state-of-the-art method, spectral hashing.

Existing unsupervised deep product quantization methods primarily aim for the increased similarity between different views of the identical image, whereas the delicate multi-level semantic similarities preserved between images are overlooked. Moreover, these methods predominantly focus on the Euclidean space for computational convenience, compromising their ability to map the multi-level semantic relationships between images effectively. To mitigate these shortcomings, we propose a novel unsupervised product quantization method dubbed \textbf{Hi}erarchical \textbf{H}yperbolic \textbf{P}roduct \textbf{Q}uantization (HiHPQ), which learns quantized representations by incorporating hierarchical semantic similarity within hyperbolic geometry. Specifically, we propose a hyperbolic product quantizer, where the hyperbolic codebook attention mechanism and the quantized contrastive learning on the hyperbolic product manifold are introduced to expedite quantization. Furthermore, we propose a hierarchical semantics learning module, designed to enhance the distinction between similar and non-matching images for a query by utilizing the extracted hierarchical semantics as an additional training supervision. Experiments on benchmarks show that our proposed method outperforms state-of-the-art baselines.

Hashing has been a widely-adopted technique for nearest neighbor search in large-scale image retrieval tasks. Recent research has shown that leveraging supervised information can lead to high quality hashing. However, the cost of annotating data is often an obstacle when applying supervised hashing to a new domain. Moreover, the results can suffer from the robustness problem as the data at training and test stage may come from different distributions. This paper studies the exploration of generating synthetic data through semisupervised generative adversarial networks (GANs), which leverages largely unlabeled and limited labeled training data to produce highly compelling data with intrinsic invariance and global coherence, for better understanding statistical structures of natural data. We demonstrate that the above two limitations can be well mitigated by applying the synthetic data for hashing. Specifically, a novel deep semantic hashing with GANs (DSH-GANs) is presented, which mainly consists of four components: a deep convolution neural networks (CNN) for learning image representations, an adversary stream to distinguish synthetic images from real ones, a hash stream for encoding image representations to hash codes and a classification stream. The whole architecture is trained endto-end by jointly optimizing three losses, i.e., adversarial loss to correct label of synthetic or real for each sample, triplet ranking loss to preserve the relative similarity ordering in the input real-synthetic triplets and classification loss to classify each sample accurately. Extensive experiments conducted on both CIFAR-10 and NUS-WIDE image benchmarks validate the capability of exploiting synthetic images for hashing. Our framework also achieves superior results when compared to state-of-the-art deep hash models.

With the recent rise in popularity and size of social media, there is a growing need for systems that can extract useful information from this amount of data. We address the problem of detecting new events from a stream of Twitter posts. To make event detection feasible on web-scale corpora, we present an algorithm based on locality-sensitive hashing which is able overcome the limitations of traditional approaches, while maintaining competitive results. In particular, a comparison with a stateof-the-art system on the first story detection task shows that we achieve over an order of magnitude speedup in processing time, while retaining comparable performance. Event detection experiments on a collection of 160 million Twitter posts show that celebrity deaths are the fastest spreading news on Twitter.

First story detection (FSD) involves identifying first stories about events from a continuous stream of documents. A major problem in this task is the high degree of lexical variation in documents which makes it very difficult to detect stories that talk about the same event but expressed using different words. We suggest using paraphrases to alleviate this problem, making this the first work to use paraphrases for FSD. We show a novel way of integrating paraphrases with locality sensitive hashing (LSH) in order to obtain an efficient FSD system that can scale to very large datasets. Our system achieves state-of-the-art results on the first story detection task, beating both the best supervised and unsupervised systems. To test our approach on large data, we construct a corpus of events for Twitter, consisting of 50 million documents, and show that paraphrasing is also beneficial in this domain.

Hashing functions, which are created to provide brief and erratic digests for the message entered, are the primary cryptographic primitives used in blockchain networks. Hashing is employed in blockchain networks to create linked block lists, which offer safe and secure distributed repository storage for critical information. Due to the unique nature of the hash search problem in blockchain networks, the most parallelization of calculations is possible. This technical report presents a performance evaluation of three popular hashing algorithms Blake3, SHA-256, and SHA-512. These hashing algorithms are widely used in various applications, such as digital signatures, message authentication, and password storage. It then discusses the performance metrics used to evaluate the algorithms, such as hash rate/throughput and memory usage. The evaluation is conducted on a range of hardware platforms, including desktop and VMs. The evaluation includes synthetic benchmarks. The results of the evaluation show that Blake3 generally outperforms both SHA-256 and SHA-512 in terms of throughput and latency. However, the performance advantage of Blake3 varies depending on the specific hardware platform and the size of the input data. The report concludes with recommendations for selecting the most suitable hashing algorithm for a given application, based on its performance requirements and security needs. The evaluation results can also inform future research and development efforts to improve the performance and security of hashing algorithms.

It is well known that high-dimensional nearest-neighbor retrieval is very expensive. Dramatic performance gains are obtained using approximate search schemes, such as the popular Locality-Sensitive Hashing (LSH). Several extensions have been proposed to address the limitations of this algorithm, in particular, by choosing more appropriate hash functions to better partition the vector space. All the proposed extensions, however, rely on a structured quantizer for hashing, poorly fitting real data sets, limiting its performance in practice. In this paper, we compare several families of space hashing functions in a real setup, namely when searching for high-dimension SIFT descriptors. The comparison of random projections, lattice quantizers, k-means and hierarchical k-means reveal that unstructured quantizer significantly improves the accuracy of LSH, as it closely fits the data in the feature space. We then compare two querying mechanisms introduced in the literature with the one originally proposed in LSH, and discuss their respective merits and limitations.

The collected documents are selected sections from the Wikipedia’s featured articles collection. This is a continuously growing dataset, that at the time of collection (October 2009) had 2,669 articles spread over 29 categories. Some of the categories are very scarce, therefore we considered only the 10 most populated ones. The articles generally have multiple sections and pictures. We have split them into sections based on section headings, and assign each image to the section in which it was placed by the author(s). Then this dataset was prunned to keep only sections that contained a single image and at least 70 words. The final corpus contains 2,866 multimedia documents. The median text length is 200 words.

Although hashing techniques have been popular for the large scale similarity search problem, most of the existing methods for designing optimal hash functions focus on homogeneous similarity assessment, i.e., the data entities to be indexed are of the same type. Realizing that heterogeneous entities and relationships are also ubiquitous in the real world applications, there is an emerging need to retrieve and search similar or relevant data entities from multiple heterogeneous domains, e.g., recommending relevant posts and images to a certain Facebook user. In this paper, we address the problem of ``comparing apples to oranges’’ under the large scale setting. Specifically, we propose a novel Relation-aware Heterogeneous Hashing (RaHH), which provides a general framework for generating hash codes of data entities sitting in multiple heterogeneous domains. Unlike some existing hashing methods that map heterogeneous data in a common Hamming space, the RaHH approach constructs a Hamming space for each type of data entities, and learns optimal mappings between them simultaneously. This makes the learned hash codes flexibly cope with the characteristics of different data domains. Moreover, the RaHH framework encodes both homogeneous and heterogeneous relationships between the data entities to design hash functions with improved accuracy. To validate the proposed RaHH method, we conduct extensive evaluations on two large datasets; one is crawled from a popular social media sites, Tencent Weibo, and the other is an open dataset of Flickr(NUS-WIDE). The experimental results clearly demonstrate that the RaHH outperforms several state-of-the-art hashing methods with significant performance gains.

With hash tables being one of the most used data structures, Lehmann, Sanders and Walzer propose a novel, light-weight hash table, referred to as Slick Hash. Their idea is to hit a sweet spot between space consumption and speed. Building on the theoretical ideas by the authors, an implementation and experiments are required to evaluate the practical performance of Slick Hash. This work contributes to fulfilling this requirement by providing a basic implementation of Slick Hash, an analysis of its performance, and an evaluation of the entry deletion, focusing on the impact of backyard cleaning. The findings are discussed, and a conclusion is drawn.

Motivated by large-scale multimedia applications we propose to learn mappings from high-dimensional data to binary codes that preserve semantic similarity. Binary codes are well suited to large-scale applications as they are storage efficient and permit exact sub-linear kNN search. The framework is applicable to broad families of mappings, and uses a flexible form of triplet ranking loss. We overcome discontinuous optimization of the discrete mappings by minimizing a piecewise-smooth upper bound on empirical loss, inspired by latent structural SVMs. We develop a new loss-augmented inference algorithm that is quadratic in the code length. We show strong retrieval performance on CIFAR-10 and MNIST, with promising classification results using no more than kNN on the binary codes.

We propose a method for learning similaritypreserving hash functions that map highdimensional data onto binary codes. The formulation is based on structured prediction with latent variables and a hinge-like loss function. It is efficient to train for large datasets, scales well to large code lengths, and outperforms state-of-the-art methods.

Existing fine-grained hashing methods typically lack code interpretability as they compute hash code bits holistically using both global and local features. To address this limitation, we propose ConceptHash, a novel method that achieves sub-code level interpretability. In ConceptHash, each sub-code corresponds to a human-understandable concept, such as an object part, and these concepts are automatically discovered without human annotations. Specifically, we leverage a Vision Transformer architecture and introduce concept tokens as visual prompts, along with image patch tokens as model inputs. Each concept is then mapped to a specific sub-code at the model output, providing natural sub-code interpretability. To capture subtle visual differences among highly similar sub-categories (e.g., bird species), we incorporate language guidance to ensure that the learned hash codes are distinguishable within fine-grained object classes while maintaining semantic alignment. This approach allows us to develop hash codes that exhibit similarity within families of species while remaining distinct from species in other families. Extensive experiments on four fine-grained image retrieval benchmarks demonstrate that ConceptHash outperforms previous methods by a significant margin, offering unique sub-code interpretability as an additional benefit. Code at: https://github.com/kamwoh/concepthash.

Unsupervised hashing methods typically aim to preserve the similarity between data points in a feature space by mapping them to binary hash codes. However, these methods often overlook the fact that the similarity between data points in the continuous feature space may not be preserved in the discrete hash code space, due to the limited similarity range of hash codes. The similarity range is bounded by the code length and can lead to a problem known as similarity collapse. That is, the positive and negative pairs of data points become less distinguishable from each other in the hash space. To alleviate this problem, in this paper a novel Simialrity Distribution Calibration (SDC) method is introduced. SDC aligns the hash code similarity distribution towards a calibration distribution (e.g., beta distribution) with sufficient spread across the entire similarity range, thus alleviating the similarity collapse problem. Extensive experiments show that our SDC outperforms significantly the state-of-the-art alternatives on coarse category-level and instance-level image retrieval.

We construct geohashing procedure based on using of space-filling H-curve. This curve provides a way to construct geohash with less computations than the construction based on usage of Hilbert curve. At the same time, H-curve has better clustering properties.

When approximating binary similarity using the hamming distance between short binary hashes, we show that even if the similarity is symmetric, we can have shorter and more accurate hashes by using two distinct code maps. I.e. by approximating the similarity between x and x 0 as the hamming distance between f(x) and g(x0), for two distinct binary codes f, g, rather than as the hamming distance between f(x) and f(x0).

The pervasiveness of massive data repositories has led to much interest in efficient methods for indexing, search, and retrieval. For image data, a rapidly developing body of work for these applications shows impressive performance with methods that broadly fall under the umbrella term of Binary Hashing. Given a distance matrix, a binary hashing algorithm solves for a binary code for the given set of examples, whose Hamming distance nicely approximates the original distances. The formulation is non-convex — so existing solutions adopt spectral relaxations or perform coordinate descent (or quantization) on a surrogate objective that is numerically more tractable. In this paper, we first derive an Augmented Lagrangian approach to optimize the standard binary Hashing objective (i.e., maintain fidelity with a given distance matrix). With appropriate step sizes, we find that this scheme already yields results that match or substantially outperform state of the art methods on most benchmarks used in the literature. Then, to allow the model to scale to large datasets, we obtain an interesting reformulation of the binary hashing objective as a non-negative matrix factorization. Later, this leads to a simple multiplicative updates algorithm — whose parallelization properties are exploited to obtain a fast GPU based implementation. We give a probabilistic analysis of our initialization scheme and present a range of experiments to show that the method is simple to implement and competes favorably with available methods (both for optimization and generalization).

Dense retrieval techniques employ pre-trained large language models to build a high-dimensional representation of queries and passages. These representations compute the relevance of a passage w.r.t. to a query using efficient similarity measures. In this line, multi-vector representations show improved effectiveness at the expense of a one-order-of-magnitude increase in memory footprint and query latency by encoding queries and documents on a per-token level. Recently, PLAID has tackled these problems by introducing a centroid-based term representation to reduce the memory impact of multi-vector systems. By exploiting a centroid interaction mechanism, PLAID filters out non-relevant documents, thus reducing the cost of the successive ranking stages. This paper proposes ``Efficient Multi-Vector dense retrieval with Bit vectors’’ (EMVB), a novel framework for efficient query processing in multi-vector dense retrieval. First, EMVB employs a highly efficient pre-filtering step of passages using optimized bit vectors. Second, the computation of the centroid interaction happens column-wise, exploiting SIMD instructions, thus reducing its latency. Third, EMVB leverages Product Quantization (PQ) to reduce the memory footprint of storing vector representations while jointly allowing for fast late interaction. Fourth, we introduce a per-document term filtering method that further improves the efficiency of the last step. Experiments on MS MARCO and LoTTE show that EMVB is up to 2.8x faster while reducing the memory footprint by 1.8x with no loss in retrieval accuracy compared to PLAID.

Image hash codes are produced by binarizing the embeddings of convolutional neural networks (CNN) trained for either classification or retrieval. While proxy embeddings achieve good performance on both tasks, they are non-trivial to binarize, due to a rotational ambiguity that encourages non-binary embeddings. The use of a fixed set of proxies (weights of the CNN classification layer) is proposed to eliminate this ambiguity, and a procedure to design proxy sets that are nearly optimal for both classification and hashing is introduced. The resulting hash-consistent large margin (HCLM) proxies are shown to encourage saturation of hashing units, thus guaranteeing a small binarization error, while producing highly discriminative hash-codes. A semantic extension (sHCLM), aimed to improve hashing performance in a transfer scenario, is also proposed. Extensive experiments show that sHCLM embeddings achieve significant improvements over state-of-the-art hashing procedures on several small and large datasets, both within and beyond the set of training classes.

In this paper we focus on improving the effectiveness of hashing-based approximate nearest neighbour search. Generating similarity preserving hashcodes for images has been shown to be an effective and efficient method for searching through large datasets. Hashcode generation generally involves two steps: bucketing the input feature space with a set of hyperplanes, followed by quantising the projection of the data-points onto the normal vectors to those hyperplanes. This procedure results in the makeup of the hashcodes depending on the positions of the data-points with respect to the hyperplanes in the feature space, allowing a degree of locality to be encoded into the hashcodes. In this paper we study the effect of learning both the hyperplanes and the thresholds as part of the same model. Most previous research either learn the hyperplanes assuming a fixed set of thresholds, or vice-versa. In our experiments over two standard image datasets we find statistically significant increases in retrieval effectiveness versus a host of state-of-the-art data-dependent and independent hashing models.

In this paper we show how word embeddings can be used to increase the effectiveness of a state-of-the art Locality Sensitive Hashing (LSH) based first story detection (FSD) system over a standard tweet corpus. Vocabulary mismatch, in which related tweets use different words, is a serious hindrance to the effectiveness of a modern FSD system. In this case, a tweet could be flagged as a first story even if a related tweet, which uses different but synonymous words, was already returned as a first story. In this work, we propose a novel approach to mitigate this problem of lexical variation, based on tweet expansion. In particular, we propose to expand tweets with semantically related paraphrases identified via automatically mined word embeddings over a background tweet corpus. Through experimentation on a large data stream comprised of 50 million tweets, we show that FSD effectiveness can be improved by 9.5% over a state-of-the-art FSD system.

In this paper we propose a two-step iterative scheme, Graph Regularised Hashing (GRH), for incrementally adjusting the positioning of the hashing hypersurfaces to better conform to the supervisory signal: in the first step the binary bits are regularised using a data similarity graph so that similar data points receive similar bits. In the second step the regularised hashcodes form targets for a set of binary classifiers which shift the position of each hypersurface so as to separate opposite bits with maximum margin. GRH exhibits superior retrieval accuracy to competing hashing methods.

We introduce a scheme for optimally allocating multiple bits per hyperplane for Locality Sensitive Hashing (LSH). Existing approaches binarise LSH projections by thresholding at zero yielding a single bit per dimension. We demonstrate that this is a sub-optimal bit allocation approach that can easily destroy the neighbourhood structure in the original feature space. Our proposed method, dubbed Neighbourhood Preserving Quantization (NPQ), assigns multiple bits per hyperplane based upon adaptively learned thresholds. NPQ exploits a pairwise affinity matrix to discretise each dimension such that nearest neighbours in the original feature space fall within the same quantisation thresholds and are therefore assigned identical bits. NPQ is not only applicable to LSH, but can also be applied to any low-dimensional projection scheme. Despite using half the number of hyperplanes, NPQ is shown to improve LSH-based retrieval accuracy by up to 65% compared to the state-of-the-art.

We explore methods to solve the multi-label classification task posed by the GeoLifeCLEF 2024 competition with the DS@GT team, which aims to predict the presence and absence of plant species at specific locations using spatial and temporal remote sensing data. Our approach uses frequency-domain coefficients via the Discrete Cosine Transform (DCT) to compress and pre-compute the raw input data for convolutional neural networks. We also investigate nearest neighborhood models via locality-sensitive hashing (LSH) for prediction and to aid in the self-supervised contrastive learning of embeddings through tile2vec. Our best competition model utilized geolocation features with a leaderboard score of 0.152 and a best post-competition score of 0.161. Source code and models are available at https://github.com/dsgt-kaggle-clef/geolifeclef-2024.

In this paper we propose Regularised Cross-Modal Hashing (RCMH) a new cross-modal hashing scheme that projects annotation and visual feature descriptors into a common Hamming space. RCMH optimises the intra-modality similarity of data-points in the annotation modality using an iterative three-step hashing algorithm: in the first step each training image is assigned a K-bit hashcode based on hyperplanes learnt at the previous iteration; in the second step the binary bits are smoothed by a formulation of graph regularisation so that similar data-points have similar bits; in the third step a set of binary classifiers are trained to predict the regularised bits with maximum margin. Visual descriptors are projected into the annotation Hamming space by a set of binary classifiers learnt using the bits of the corresponding annotations as labels. RCMH is shown to consistently improve retrieval effectiveness over state-of-the-art baselines.

This study introduces a novel transformer model optimized for large-scale point cloud processing in scientific domains such as high-energy physics (HEP) and astrophysics. Addressing the limitations of graph neural networks and standard transformers, our model integrates local inductive bias and achieves near-linear complexity with hardware-friendly regular operations. One contribution of this work is the quantitative analysis of the error-complexity tradeoff of various sparsification techniques for building efficient transformers. Our findings highlight the superiority of using locality-sensitive hashing (LSH), especially OR & AND-construction LSH, in kernel approximation for large-scale point cloud data with local inductive bias. Based on this finding, we propose LSH-based Efficient Point Transformer (HEPT), which combines E\(^2\)LSH with OR & AND constructions and is built upon regular computations. HEPT demonstrates remarkable performance on two critical yet time-consuming HEP tasks, significantly outperforming existing GNNs and transformers in accuracy and computational speed, marking a significant advancement in geometric deep learning and large-scale scientific data processing. Our code is available at https://github.com/Graph-COM/HEPT.

Microsoft SPACEV-1B is a new web search related dataset released by Microsoft Bing for this competition. It consists of document and query vectors encoded by Microsoft SpaceV Superior model to capture generic intent representation.

Approximate nearest neighbor search (ANN) data structures have widespread applications in machine learning, computational biology, and text processing. The goal of ANN is to preprocess a set S so that, given a query q, we can find a point y whose distance from q approximates the smallest distance from q to any point in S. For most distance functions, the best-known ANN bounds for high-dimensional point sets are obtained using techniques based on locality-sensitive hashing (LSH). Unfortunately, space efficiency is a major challenge for LSH-based data structures. Classic LSH techniques require a very large amount of space, oftentimes polynomial in |S|. A long line of work has developed intricate techniques to reduce this space usage, but these techniques suffer from downsides: they must be hand tailored to each specific LSH, are often complicated, and their space reduction comes at the cost of significantly increased query times. In this paper we explore a new way to improve the space efficiency of LSH using function inversion techniques, originally developed in (Fiat and Naor 2000). We begin by describing how function inversion can be used to improve LSH data structures. This gives a fairly simple, black box method to reduce LSH space usage. Then, we give a data structure that leverages function inversion to improve the query time of the best known near-linear space data structure for approximate nearest neighbor search under Euclidean distance: the ALRW data structure of (Andoni, Laarhoven, Razenshteyn, and Waingarten 2017). ALRW was previously shown to be optimal among “list-of-points” data structures for both Euclidean and Manhattan ANN; thus, in addition to giving improved bounds, our results imply that list-of-points data structures are not optimal for Euclidean or Manhattan ANN.

This paper examines the graph partition problem and introduces a new metric, MSIDS (maximal sum of inner degrees squared). We establish its connection to the replication factor (RF) optimization, which has been the main focus of theoretical work in this field. Additionally, we propose a new partition algorithm, DBH-X, based on the DBH partitioner. We demonstrate that DBH-X significantly improves both the RF and MSIDS, compared to the baseline DBH algorithm. In addition, we provide test results that show the runtime acceleration of GraphX-based PageRank and Label propagation algorithms.

The advent of the Internet of Things (IoT) has brought forth additional intricacies and difficulties to computer networks. These gadgets are particularly susceptible to cyber-attacks because of their simplistic design. Therefore, it is crucial to recognise these devices inside a network for the purpose of network administration and to identify any harmful actions. Network traffic fingerprinting is a crucial technique for identifying devices and detecting anomalies. Currently, the predominant methods for this depend heavily on machine learning (ML). Nevertheless, machine learning (ML) methods need the selection of features, adjustment of hyperparameters, and retraining of models to attain optimal outcomes and provide resilience to concept drifts detected in a network. In this research, we suggest using locality-sensitive hashing (LSH) for network traffic fingerprinting as a solution to these difficulties. Our study focuses on examining several design options for the Nilsimsa LSH function. We then use this function to create unique fingerprints for network data, which may be used to identify devices. We also compared it with ML-based traffic fingerprinting and observed that our method increases the accuracy of state-of-the-art by 12% achieving around 94% accuracy in identifying devices in a network.

Consistent range-hashing is a technique used in distributed systems, either directly or as a subroutine for consistent hashing, commonly to realize an even and stable data distribution over a variable number of resources. We introduce FlipHash, a consistent range-hashing algorithm with constant time complexity and low memory requirements. Like Jump Consistent Hash, FlipHash is intended for applications where resources can be indexed sequentially. Under this condition, it ensures that keys are hashed evenly across resources and that changing the number of resources only causes keys to be remapped from a removed resource or to an added one, but never shuffled across persisted ones. FlipHash differentiates itself with its low computational cost, achieving constant-time complexity. We show that FlipHash beats Jump Consistent Hash’s cost, which is logarithmic in the number of resources, both theoretically and in experiments over practical settings.

We introduce a scheme for optimally allocating a variable number of bits per LSH hyperplane. Previous approaches assign a constant number of bits per hyperplane. This neglects the fact that a subset of hyperplanes may be more informative than others. Our method, dubbed Variable Bit Quantisation (VBQ), provides a datadriven non-uniform bit allocation across hyperplanes. Despite only using a fraction of the available hyperplanes, VBQ outperforms uniform quantisation by up to 168% for retrieval across standard text and image datasets.

Recent years have witnessed the success of the emerging hashing techniques in large-scale image retrieval. Owing to the great learning capacity, deep hashing has become one of the most promising solutions, and achieved attractive performance in practice. However, without semantic label information, the unsupervised deep hashing still remains an open question. In this paper, we propose a novel progressive generative hashing (PGH) framework to help learn a discriminative hashing network in an unsupervised way. Different from existing studies, it first treats the hash codes as a kind of semantic condition for the similar image generation, and simultaneously feeds the original image and its codes into the generative adversarial networks (GANs). The real images together with the synthetic ones can further help train a discriminative hashing network based on a triplet loss. By iteratively inputting the learnt codes into the hash conditioned GANs, we can progressively enable the hashing network to discover the semantic relations. Extensive experiments on the widely-used image datasets demonstrate that PGH can significantly outperform stateof-the-art unsupervised hashing methods.

This article studies deep unsupervised hashing, which has attracted increasing attention in large-scale image retrieval. The majority of recent approaches usually reconstruct semantic similarity information, which then guides the hash code learning. However, they still fail to achieve satisfactory performance in reality for two reasons. On the one hand, without accurate supervised information, these methods usually fail to produce independent and robust hash codes with semantics information well preserved, which may hinder effective image retrieval. On the other hand, due to discrete constraints, how to effectively optimize the hashing network in an end-to-end manner with small quantization errors remains a problem. To address these difficulties, we propose a novel unsupervised hashing method called HARR to learn discriminative and high-quality hash codes. To comprehensively explore semantic similarity structure, HARR adopts the Winner-Take-All hash to model the similarity structure. Then similarity-preserving hash codes are learned under the reliable guidance of the reconstructed similarity structure. Additionally, we improve the quality of hash codes by a bit correlation reduction module, which forces the cross-correlation matrix between a batch of hash codes under different augmentations to approach the identity matrix. In this way, the generated hash bits are expected to be invariant to disturbances with minimal redundancy, which can be further interpreted as an instantiation of the information bottleneck principle. Finally, for effective hashing network training, we minimize the cosine distances between real-value network outputs and their binary codes for small quantization errors. Extensive experiments demonstrate the effectiveness of our proposed HARR.

Spectral Toolkit of Algorithms for Graphs (STAG) is an open-source library for efficient graph algorithms. This technical report presents the newly implemented component on locality sensitive hashing, kernel density estimation, and fast spectral clustering. The report includes a user’s guide to the newly implemented algorithms, experiments and demonstrations of the new functionality, and several technical considerations behind our development.

Despite significant progress in supervised hashing, there are three common limitations of existing methods. First, most pioneer methods discretely learn hash codes bit by bit, making the learning procedure rather time-consuming. Second, to reduce the large complexity of the n by n pairwise similarity matrix, most methods apply sampling strategies during training, which inevitably results in information loss and suboptimal performance; some recent methods try to replace the large matrix with a smaller one, but the size is still large. Third, among the methods that leverage the pairwise similarity matrix, most of them only encode the semantic label information in learning the hash codes, failing to fully capture the characteristics of data. In this paper, we present a novel supervised hashing method, called Fast Scalable Supervised Hashing (FSSH), which circumvents the use of the large similarity matrix by introducing a pre-computed intermediate term whose size is independent with the size of training data. Moreover, FSSH can learn the hash codes with not only the semantic information but also the features of data. Extensive experiments on three widely used datasets demonstrate its superiority over several state-of-the-art methods in both accuracy and scalability. Our experiment codes are available at: https://lcbwlx.wixsite.com/fssh.

Huge embedding tables in modern Deep Learning Recommender Models (DLRM) require prohibitively large memory during training and inference. Aiming to reduce the memory footprint of training, this paper proposes FIne-grained In-Training Embedding Dimension optimization (FIITED). Given the observation that embedding vectors are not equally important, FIITED adjusts the dimension of each individual embedding vector continuously during training, assigning longer dimensions to more important embeddings while adapting to dynamic changes in data. A novel embedding storage system based on virtually-hashed physically-indexed hash tables is designed to efficiently implement the embedding dimension adjustment and effectively enable memory saving. Experiments on two industry models show that FIITED is able to reduce the size of embeddings by more than 65% while maintaining the trained model’s quality, saving significantly more memory than a state-of-the-art in-training embedding pruning method. On public click-through rate prediction datasets, FIITED is able to prune up to 93.75%-99.75% embeddings without significant accuracy loss.

Multi-modal hashing is an effective technique to support large-scale multimedia retrieval, due to its capability of encoding heterogeneous multi-modal features into compact and similarity-preserving binary codes. Although great progress has been achieved so far, existing methods still suffer from several problems, including: 1) All existing methods simply adopt fixed modality combination weights in online hashing process to generate the query hash codes. This strategy cannot adaptively capture the variations of different queries. 2) They either suffer from insufficient semantics (for unsupervised methods) or require high computation and storage cost (for the supervised methods, which rely on pair-wise semantic matrix). 3) They solve the hash codes with relaxed optimization strategy or bit-by-bit discrete optimization, which results in significant quantization loss or consumes considerable computation time. To address the above limitations, in this paper, we propose an Online Multi-modal Hashing with Dynamic Query-adaption (OMH-DQ) method in a novel fashion. Specifically, a self-weighted fusion strategy is designed to adaptively preserve the multi-modal feature information into hash codes by exploiting their complementarity. The hash codes are learned with the supervision of pair-wise semantic labels to enhance their discriminative capability, while avoiding the challenging symmetric similarity matrix factorization. Under such learning framework, the binary hash codes can be directly obtained with efficient operations and without quantization errors. Accordingly, our method can benefit from the semantic labels, and simultaneously, avoid the high computation complexity. Moreover, to accurately capture the query variations, at the online retrieval stage, we design a parameter-free online hashing module which can adaptively learn the query hash codes according to the dynamic query contents. Extensive experiments demonstrate the state-of-the-art performance of the proposed approach from various aspects.

Hashing has attracted widespread attention in image retrieval because of its fast retrieval speed and low storage cost. Compared with supervised methods, unsupervised hashing methods are more reasonable and suitable for large-scale image retrieval since it is always difficult and expensive to collect true labels of the massive data. Without label information, however, unsupervised hashing methods can not guarantee the quality of learned binary codes. To resolve this dilemma, this paper proposes a novel unsupervised hashing method called Label Self-Adaption Hashing (LSAH), which contains effective hashing function learning part and self-adaption label generation part. In the first part, we utilize anchor graph to keep the local structure of the data and introduce joint sparsity into the model to extract effective features for high-quality binary code learning. In the second part, a self-adaptive cluster label matrix is learned from the data under the assumption that the nearest neighbor points should have a large probability to be in the same cluster. Therefore, the proposed LSAH can make full use of the potential discriminative information of the data to guide the learning of binary code. It is worth noting that LSAH can learn effective binary codes, hashing function and cluster labels simultaneously in a unified optimization framework. To solve the resulting optimization problem, an Augmented Lagrange Multiplier based iterative algorithm is elaborately designed. Extensive experiments on three large-scale data sets indicate the promising performance of the proposed LSAH.

Nearest neighbor search aims at obtaining the samples in the database with the smallest distances from them to the queries, which is a basic task in a range of fields, including computer vision and data mining. Hashing is one of the most widely used methods for its computational and storage efficiency. With the development of deep learning, deep hashing methods show more advantages than traditional methods. In this survey, we detailedly investigate current deep hashing algorithms including deep supervised hashing and deep unsupervised hashing. Specifically, we categorize deep supervised hashing methods into pairwise methods, ranking-based methods, pointwise methods as well as quantization according to how measuring the similarities of the learned hash codes. Moreover, deep unsupervised hashing is categorized into similarity reconstruction-based methods, pseudo-label-based methods, and prediction-free self-supervised learning-based methods based on their semantic learning manners. We also introduce three related important topics including semi-supervised deep hashing, domain adaption deep hashing, and multi-modal deep hashing. Meanwhile, we present some commonly used public datasets and the scheme to measure the performance of deep hashing algorithms. Finally, we discuss some potential research directions in conclusion.

The recent advances in deep neural networks have demonstrated high capability in a wide variety of scenarios. Nevertheless, fine-tuning deep models in a new domain still requires a significant amount of labeled data despite expensive labeling efforts. A valid question is how to leverage the source knowledge plus unlabeled or only sparsely labeled target data for learning a new model in target domain. The core problem is to bring the source and target distributions closer in the feature space. In the paper, we facilitate this issue in an adversarial learning framework, in which a domain discriminator is devised to handle domain shift. Particularly, we explore the learning in the context of hashing problem, which has been studied extensively due to its great efficiency in gigantic data. Specifically, a novel Deep Domain Adaptation Hashing with Adversarial learning (DeDAHA) architecture is presented, which mainly consists of three components: a deep convolutional neural networks (CNN) for learning basic image/frame representation followed by an adversary stream on one hand to optimize the domain discriminator, and on the other, to interact with each domain-specific hashing stream for encoding image representation to hash codes. The whole architecture is trained end-to-end by jointly optimizing two types of losses, i.e., triplet ranking loss to preserve the relative similarity ordering in the input triplets and adversarial loss to maximally fool the domain discriminator with the learnt source and target feature distributions. Extensive experiments are conducted on three domain transfer tasks, including cross-domain digits retrieval, image to image and image to video transfers, on several benchmarks. Our DeDAHA framework achieves superior results when compared to the state-of-the-art techniques.

Recent years have witnessed the growing popularity of hashing in large-scale vision problems. It has been shown that the hashing quality could be boosted by leveraging supervised information into hash function learning. However, the existing supervised methods either lack adequate performance or often incur cumbersome model training. In this paper, we propose a novel kernel-based supervised hashing model which requires a limited amount of supervised information, i.e., similar and dissimilar data pairs, and a feasible training cost in achieving high quality hashing. The idea is to map the data to compact binary codes whose Hamming distances are minimized on similar pairs and simultaneously maximized on dissimilar pairs. Our approach is distinct from prior works by utilizing the equivalence between optimizing the code inner products and the Hamming distances. This enables us to sequentially and efficiently train the hash functions one bit at a time, yielding very short yet discriminative codes. We carry out extensive experiments on two image benchmarks with up to one million samples, demonstrating that our approach significantly outperforms the state-of-the-arts in searching both metric distance neighbors and semantically similar neighbors, with accuracy gains ranging from 13% to 46%.

Efficient hashing techniques have attracted extensive research interests in both storage and retrieval of high dimensional data, such as images and videos. In existing hashing methods, a linear model is commonly utilized owing to its efficiency. To obtain better accuracy, linear-based hashing methods focus on designing a generalized linear objective function with different constraints or penalty terms that consider the inherent characteristics and neighborhood information of samples. Differing from existing hashing methods, in this study, we propose a self-improvement framework called Model Boost (MoBoost) to improve model parameter optimization for linear-based hashing methods without adding new constraints or penalty terms. In the proposed MoBoost, for a linear-based hashing method, we first repeatedly execute the hashing method to obtain several hash codes to training samples. Then, utilizing two novel fusion strategies, these codes are fused into a single set. We also propose two new criteria to evaluate the goodness of hash bits during the fusion process. Based on the fused set of hash codes, we learn new parameters for the linear hash function that can significantly improve the accuracy. In general, the proposed MoBoost can be adopted by existing linear-based hashing methods, achieving more precise and stable performance compared to the original methods, and adopting the proposed MoBoost will incur negligible time and space costs. To evaluate the proposed MoBoost, we performed extensive experiments on four benchmark datasets, and the results demonstrate superior performance.

The linear model is commonly utilized in hashing methods owing to its efficiency. To obtain better accuracy, linear-based hashing methods focus on designing a generalized linear objective function with different constraints or penalty terms that consider neighborhood information. In this study, we propose a novel generalized framework called Model Boost (MoBoost), which can achieve the self-improvement of the linear-based hashing. The proposed MoBoost is used to improve model parameter optimization for linear-based hashing methods without adding new constraints or penalty terms. In the proposed MoBoost, given a linear-based hashing method, we first execute the method several times to get several different hash codes for training samples, and then combine these different hash codes into one set utilizing one novel fusion strategy. Based on this set of hash codes, we learn some new parameters for the linear hash function that can significantly improve accuracy. The proposed MoBoost can be generally adopted in existing linear-based hashing methods, achieving more precise and stable performance compared to the original methods while imposing negligible added expenditure in terms of time and space. Extensive experiments are performed based on three benchmark datasets, and the results demonstrate the superior performance of the proposed framework.

Recent years have witnessed the success of hashing techniques in fast nearest neighbor search. In practice many applications (e.g., visual search, object detection, image matching, etc.) have enjoyed the benefits of complementary hash tables and information fusion over multiple views. However, most of prior research mainly focused on compact hash code cleaning, and rare work studies how to build multiple complementary hash tables, much less to adaptively integrate information stemming from multiple views. In this paper we first present a novel multi-view complementary hash table method that learns complementary hash tables from the data with multiple views. For single multiview table, using exemplar based feature fusion, we approximate the inherent data similarities with a low-rank matrix, and learn discriminative hash functions in an efficient way. To build complementary tables and meanwhile maintain scalable training and fast out-of-sample extension, an exemplar reweighting scheme is introduced to update the induced low-rank similarity in the sequential table construction framework, which indeed brings mutual benefits between tables by placing greater importance on exemplars shared by mis-separated neighbors. Extensive experiments on three large-scale image datasets demonstrate that the proposed method significantly outperforms various naive solutions and state-of-the-art multi-table methods.

Hashing based methods recently have been shown promising for large-scale nearest neighbor search. However, good designs involve difficult decisions of many unknowns – data features, hashing algorithms, parameter settings, kernels, etc. In this paper, we provide a unified solution as hash bit selection, i.e., selecting the most informative hash bits from a pool of candidates that may have been generated under various conditions mentioned above. We represent the candidate bit pool as a vertex- and edge-weighted graph with the pooled bits as vertices. Then we formulate the bit selection problem as quadratic programming over the graph, and solve it efficiently by replicator dynamics. Extensive experiments show that our bit selection approach can achieve superior performance over both naive selection methods and state-of-the-art methods under each scenario, usually with significant accuracy gains from 10% to 50% relatively.

Learning to hash has been recognized to accomplish highly efficient storage and retrieval for large-scale visual data. Particularly, ranking-based hashing techniques have recently attracted broad research attention because ranking accuracy among the retrieved data is well explored and their objective is more applicable to realistic search tasks. However, directly optimizing discrete hash codes without continuous-relaxations on a nonlinear ranking objective is infeasible by either traditional optimization methods or even recent discrete hashing algorithms. To address this challenging issue, in this paper, we introduce a novel supervised hashing method, dubbed Discrete Semantic Ranking Hashing (DSeRH), which aims to directly embed semantic rank orders into binary codes. In DSeRH, a generalized Adaptive Discrete Minimization (ADM) approach is proposed to discretely optimize binary codes with the quadratic nonlinear ranking objective in an iterative manner and is guaranteed to converge quickly. Additionally, instead of using 0/1 independent labels to form rank orders as in previous works, we generate the listwise rank orders from the high-level semantic word embeddings which can quantitatively capture the intrinsic correlation between different categories. We evaluate our DSeRH, coupled with both linear and deep convolutional neural network (CNN) hash functions, on three image datasets, i.e., CIFAR-10, SUN397 and ImageNet100, and the results manifest that DSeRH can outperform the state-of-the-art ranking-based hashing methods.

Hashing is becoming increasingly popular for efficient nearest neighbor search in massive databases. However, learning short codes that yield good search performance is still a challenge. Moreover, in many cases realworld data lives on a low-dimensional manifold, which should be taken into account to capture meaningful nearest neighbors. In this paper, we propose a novel graph-based hashing method which automatically discovers the neighborhood structure inherent in the data to learn appropriate compact codes. To make such an approach computationally feasible, we utilize Anchor Graphs to obtain tractable low-rank adjacency matrices. Our formulation allows constant time hashing of a new data point by extrapolating graph Laplacian eigenvectors to eigenfunctions. Finally, we describe a hierarchical threshold learning procedure in which each eigenfunction yields multiple bits, leading to higher search accuracy. Experimental comparison with the other state-of-the-art methods on two large datasets demonstrates the efficacy of the proposed method.

Hashing technique has become a promising approach for fast similarity search. Most of existing hashing research pursue the binary codes for the same type of entities by preserving their similarities. In practice, there are many scenarios involving nearest neighbor search on the data given in matrix form, where two different types of, yet naturally associated entities respectively correspond to its two dimensions or views. To fully explore the duality between the two views, we propose a collaborative hashing scheme for the data in matrix form to enable fast search in various applications such as image search using bag of words and recommendation using user-item ratings. By simultaneously preserving both the entity similarities in each view and the interrelationship between views, our collaborative hashing effectively learns the compact binary codes and the explicit hash functions for out-of-sample extension in an alternating optimization way. Extensive evaluations are conducted on three well-known datasets for search inside a single view and search across different views, demonstrating that our proposed method outperforms state-of-the-art baselines, with significant accuracy gains ranging from 7.67% to 45.87% relatively.

Hashing has emerged as a popular technique for fast nearest neighbor search in gigantic databases. In particular, learning based hashing has received considerable attention due to its appealing storage and search efficiency. However, the performance of most unsupervised learning based hashing methods deteriorates rapidly as the hash code length increases. We argue that the degraded performance is due to inferior optimization procedures used to achieve discrete binary codes. This paper presents a graph-based unsupervised hashing model to preserve the neighborhood structure of massive data in a discrete code space. We cast the graph hashing problem into a discrete optimization framework which directly learns the binary codes. A tractable alternating maximization algorithm is then proposed to explicitly deal with the discrete constraints, yielding high-quality codes to well capture the local neighborhoods. Extensive experiments performed on four large datasets with up to one million samples show that our discrete optimization based graph hashing method obtains superior search accuracy over state-of-the-art unsupervised hashing methods, especially for longer codes.

Hashing-based cross-modal search which aims to map multiple modality features into binary codes has attracted increasingly attention due to its storage and search efficiency especially in large-scale database retrieval. Recent unsupervised deep cross-modal hashing methods have shown promising results. However, existing approaches typically suffer from two limitations: (1) They usually learn cross-modal similarity information separately or in a redundant fusion manner, which may fail to capture semantic correlations among instances from different modalities sufficiently and effectively. (2) They seldom consider the sampling and weighting schemes for unsupervised cross-modal hashing, resulting in the lack of satisfactory discriminative ability in hash codes. To overcome these limitations, we propose a novel unsupervised deep cross-modal hashing method called Joint-modal Distribution-based Similarity Hashing (JDSH) for large-scale cross-modal retrieval. Firstly, we propose a novel cross-modal joint-training method by constructing a joint-modal similarity matrix to fully preserve the cross-modal semantic correlations among instances. Secondly, we propose a sampling and weighting scheme termed the Distribution-based Similarity Decision and Weighting (DSDW) method for unsupervised cross-modal hashing, which is able to generate more discriminative hash codes by pushing semantic similar instance pairs closer and pulling semantic dissimilar instance pairs apart. The experimental results demonstrate the superiority of JDSH compared with several unsupervised cross-modal hashing methods on two public datasets NUS-WIDE and MIRFlickr.

Similarity indices for high-dimensional data are very desirable for building content-based search systems for featurerich data such as audio, images, videos, and other sensor data. Recently, locality sensitive hashing (LSH) and its variations have been proposed as indexing techniques for approximate similarity search. A significant drawback of these approaches is the requirement for a large number of hash tables in order to achieve good search quality. This paper proposes a new indexing scheme called multi-probe LSH that overcomes this drawback. Multi-probe LSH is built on the well-known LSH technique, but it intelligently probes multiple buckets that are likely to contain query results in a hash table. Our method is inspired by and improves upon recent theoretical work on entropy-based LSH designed to reduce the space requirement of the basic LSH method. We have implemented the multi-probe LSH method and evaluated the implementation with two different high-dimensional datasets. Our evaluation shows that the multi-probe LSH method substantially improves upon previously proposed methods in both space and time efficiency. To achieve the same search quality, multi-probe LSH has a similar timeefficiency as the basic LSH method while reducing the number of hash tables by an order of magnitude. In comparison with the entropy-based LSH method, to achieve the same search quality, multi-probe LSH uses less query time and 5 to 8 times fewer number of hash tables.

In this paper, we propose a deep variational and structural hashing (DVStH) method to learn compact binary codes for multimedia retrieval. Unlike most existing deep hashing methods which use a series of convolution and fully-connected layers to learn binary features, we develop a probabilistic framework to infer latent feature representation inside the network. Then, we design a struct layer rather than a bottleneck hash layer, to obtain binary codes through a simple encoding procedure. By doing these, we are able to obtain binary codes discriminatively and generatively. To make it applicable to cross-modal scalable multimedia retrieval, we extend our method to a cross-modal deep variational and structural hashing (CM-DVStH). We design a deep fusion network with a struct layer to maximize the correlation between image-text input pairs during the training stage so that a unified binary vector can be obtained. We then design modality-specific hashing networks to handle the out-of-sample extension scenario. Specifically, we train a network for each modality which outputs a latent representation that is as close as possible to the binary codes which are inferred from the fusion network. Experimental results on five benchmark datasets are presented to show the efficacy of the proposed approach.

In this paper, we propose a cross-modal deep variational hashing (CMDVH) method for cross-modality multimedia retrieval. Unlike existing cross-modal hashing methods which learn a single pair of projections to map each example as a binary vector, we design a couple of deep neural network to learn non-linear transformations from imagetext input pairs, so that unified binary codes can be obtained. We then design the modality-specific neural networks in a probabilistic manner where we model a latent variable as close as possible from the inferred binary codes, which is approximated by a posterior distribution regularized by a known prior. Experimental results on three benchmark datasets show the efficacy of the proposed approach.

With benefits of low storage costs and high query speeds, hashing methods are widely researched for efficiently retrieving large-scale data, which commonly contains multiple views, e.g. a news report with images, videos and texts. In this paper, we study the problem of cross-view retrieval and propose an effective Semantics-Preserving Hashing method, termed SePH. Given semantic affinities of training data as supervised information, SePH transforms them into a probability distribution and approximates it with tobe-learnt hash codes in Hamming space via minimizing the Kullback-Leibler divergence. Then kernel logistic regression with a sampling strategy is utilized to learn the nonlinear projections from features in each view to the learnt hash codes. And for any unseen instance, predicted hash codes and their corresponding output probabilities from observed views are utilized to determine its unified hash code, using a novel probabilistic approach. Extensive experiments conducted on three benchmark datasets well demonstrate the effectiveness and reasonableness of SePH.

Approximate nearest neighbor search is an efficient strategy for large-scale image retrieval. Encouraged by the recent advances in convolutional neural networks (CNNs), we propose an effective deep learning framework to generate binary hash codes for fast image retrieval. Our idea is that when the data labels are available, binary codes can be learned by employing a hidden layer for representing the latent concepts that dominate the class labels. he utilization of the CNN also allows for learning image representations. Unlike other supervised methods that require pair-wised inputs for binary code learning, our method learns hash codes and image representations in a point-wised manner, making it suitable for large-scale datasets. Experimental results show that our method outperforms several state-of-the-art hashing algorithms on the CIFAR-10 and MNIST datasets. We further demonstrate its scalability and efficacy on a large-scale dataset of 1 million clothing images.

Supervised hashing aims to map the original features to compact binary codes that are able to preserve label based similarity in the Hamming space. Non-linear hash functions have demonstrated their advantage over linear ones due to their powerful generalization capability. In the literature, kernel functions are typically used to achieve non-linearity in hashing, which achieve encouraging retrieval performance at the price of slow evaluation and training time. Here we propose to use boosted decision trees for achieving non-linearity in hashing, which are fast to train and evaluate, hence more suitable for hashing with high dimensional data. In our approach, we first propose sub-modular formulations for the hashing binary code inference problem and an efficient GraphCut based block search method for solving large-scale inference. Then we learn hash functions by training boosted decision trees to fit the binary codes. Experiments demonstrate that our proposed method significantly outperforms most state-of-the-art methods in retrieval precision and training time. Especially for highdimensional data, our method is orders of magnitude faster than many methods in terms of training time.

Most existing approaches to hashing apply a single form of hash function, and an optimization process which is typically deeply coupled to this specific form. This tight coupling restricts the flexibility of the method to respond to the data, and can result in complex optimization problems that are difficult to solve. Here we propose a flexible yet simple framework that is able to accommodate different types of loss functions and hash functions. This framework allows a number of existing approaches to hashing to be placed in context, and simplifies the development of new problem-specific hashing methods. Our framework decomposes hashing learning problem into two steps: hash bit learning and hash function learning based on the learned bits. The first step can typically be formulated as binary quadratic problems, and the second step can be accomplished by training standard binary classifiers. Both problems have been extensively studied in the literature. Our extensive experiments demonstrate that the proposed framework is effective, flexible and outperforms the state-of-the-art.

We present a new dataset with the goal of advancing the state-of-the-art in object recognition by placing the question of object recognition in the context of the broader question of scene understanding. This is achieved by gathering images of complex everyday scenes containing common objects in their natural context. Objects are labeled using per-instance segmentations to aid in precise object localization. Our dataset contains photos of 91 objects types that would be easily recognizable by a 4 year old. With a total of 2.5 million labeled instances in 328k images, the creation of our dataset drew upon extensive crowd worker involvement via novel user interfaces for category detection, instance spotting and instance segmentation. We present a detailed statistical analysis of the dataset in comparison to PASCAL, ImageNet, and SUN. Finally, we provide baseline performance analysis for bounding box and segmentation detection results using a Deformable Parts Model.

Current massive datasets demand light-weight access for analysis. Discrete hashing methods are thus beneficial because they map high-dimensional data to compact binary codes that are efficient to store and process, while preserving semantic similarity. To optimize powerful deep learning methods for image hashing, gradient-based methods are required. Binary codes, however, are discrete and thus have no continuous derivatives. Relaxing the problem by solving it in a continuous space and then quantizing the solution is not guaranteed to yield separable binary codes. The quantization needs to be included in the optimization. In this paper we push for quantization: We optimize maximum class separability in the binary space. We introduce a margin on distances between dissimilar image pairs as measured in the binary space. In addition to pair-wise distances, we draw inspiration from Fisher’s Linear Discriminant Analysis (Fisher LDA) to maximize the binary distances between classes and at the same time minimize the binary distance of images within the same class. Experiments on CIFAR-10, NUS-WIDE and ImageNet100 demonstrate compact codes comparing favorably to the current state of the art.

Embedding techniques have become essential components of large databases in the deep learning era. By encoding discrete entities, such as words, items, or graph nodes, into continuous vector spaces, embeddings facilitate more efficient storage, retrieval, and processing in large databases. Especially in the domain of recommender systems, millions of categorical features are encoded as unique embedding vectors, which facilitates the modeling of similarities and interactions among features. However, numerous embedding vectors can result in significant storage overhead. In this paper, we aim to compress the embedding table through quantization techniques. Given that features vary in importance levels, we seek to identify an appropriate precision for each feature to balance model accuracy and memory usage. To this end, we propose a novel embedding compression method, termed Mixed-Precision Embeddings (MPE). Specifically, to reduce the size of the search space, we first group features by frequency and then search precision for each feature group. MPE further learns the probability distribution over precision levels for each feature group, which can be used to identify the most suitable precision with a specially designed sampling strategy. Extensive experiments on three public datasets demonstrate that MPE significantly outperforms existing embedding compression methods. Remarkably, MPE achieves about 200x compression on the Criteo dataset without comprising the prediction accuracy.

Fast nearest neighbor searching is becoming an increasingly important tool in solving many large-scale problems. Recently a number of approaches to learning datadependent hash functions have been developed. In this work, we propose a column generation based method for learning datadependent hash functions on the basis of proximity comparison information. Given a set of triplets that encode the pairwise proximity comparison information, our method learns hash functions that preserve the relative comparison relationships in the data as well as possible within the large-margin learning framework. The learning procedure is implemented using column generation and hence is named CGHash. At each iteration of the column generation procedure, the best hash function is selected. Unlike most other hashing methods, our method generalizes to new data points naturally; and has a training objective which is convex, thus ensuring that the global optimum can be identi- fied. Experiments demonstrate that the proposed method learns compact binary codes and that its retrieval performance compares favorably with state-of-the-art methods when tested on a few benchmark datasets.

In this paper, we propose a Neighborhood Preserving Hashing (NPH) method for scalable video retrieval in an unsupervised manner. Unlike most existing deep video hashing methods which indiscriminately compress an entire video into a binary code, we embed the spatial-temporal neighborhood information into the encoding network such that the neighborhood-relevant visual content of a video can be preferentially encoded into a binary code under the guidance of the neighborhood information. Specifically, we propose a neighborhood attention mechanism which focuses on partial useful content of each input frame conditioned on the neighborhood information. We then integrate the neighborhood attention mechanism into an RNN-based reconstruction scheme to encourage the binary codes to capture the spatial-temporal structure in a video which is consistent with that in the neighborhood. As a consequence, the learned hashing functions can map similar videos to similar binary codes. Extensive experiments on three widely-used benchmark datasets validate the effectiveness of our proposed approach.

Most existing unsupervised video hashing methods are built on unidirectional models with less reliable training objectives, which underuse the correlations among frames and the similarity structure between videos. To enable efficient scalable video retrieval, we propose a self-supervised video Hashing method based on Bidirectional Transformers (BTH). Based on the encoder-decoder structure of transformers, we design a visual cloze task to fully exploit the bidirectional correlations between frames. To unveil the similarity structure between unlabeled video data, we further develop a similarity reconstruction task by establishing reliable and effective similarity connections in the video space. Furthermore, we develop a cluster assignment task to exploit the structural statistics of the whole dataset such that more discriminative binary codes can be learned. Extensive experiments implemented on three public benchmark datasets, FCVID, ActivityNet and YFCC, demonstrate the superiority of our proposed approach.

We develop 0-bit consistent weighted sampling (CWS) for efficiently estimating min-max kernel, which is a generalization of the resemblance kernel originally designed for binary data. Because the estimator of 0-bit CWS constitutes a positive definite kernel, this method can be naturally applied to large-scale data mining problems. Basically, if we feed the sampled data from 0-bit CWS to a highly efficient linear classifier (e.g., linear SVM), we effectively (and approximately) train a nonlinear classifier based on the min-max kernel. The accuracy improves as we increase the sample size.

In this paper, we first demonstrate, through an extensive classification study using kernel machines, that the min-max kernel often provides an effective measure of similarity for nonnegative data. This helps justify the use of min-max kernel. However, as the min-max kernel is nonlinear and might be difficult to be used for industrial applications with massive data, we propose to linearize the min-max kernel via 0-bit CWS, a simplification of the original CWS method.

The previous remarkable work on consistent weighted sampling (CWS) produces samples in the form of (i, t) where the i* records the location (and in fact also the weights) information analogous to the samples produced by classical minwise hashing on binary data. Because the t* is theoretically unbounded, it was not immediately clear how to effectively implement CWS for building large-scale linear classifiers. We provide a simple solution by discarding t* (which we refer to as the “0-bit” scheme). Via an extensive empirical study, we show that this 0-bit scheme does not lose essential information. We then apply 0-bit CWS for building linear classifiers to approximate min-max kernel classifiers, as extensively validated on a wide range of public datasets.

We expect this work will generate interests among data mining practitioners who would like to efficiently utilize the nonlinear information of non-binary and nonnegative data.

Zero-shot hashing (ZSH) has shown excellent success owing to its efficiency and generalization in large-scale retrieval scenarios. While considerable success has been achieved, there still exist urgent limitations. Existing works ignore the locality relationships of representations and attributes, which have effective transferability between seeable classes and unseeable classes. Also, the continuous-value attributes are not fully harnessed. In response, we conduct a COMprehensive Attribute Exploration for ZSH, named COMAE, which depicts the relationships from seen classes to unseen ones through three meticulously designed explorations, i.e., point-wise, pair-wise and class-wise consistency constraints. By regressing attributes from the proposed attribute prototype network, COMAE learns the local features that are relevant to the visual attributes. Then COMAE utilizes contrastive learning to comprehensively depict the context of attributes, rather than instance-independent optimization. Finally, the class-wise constraint is designed to cohesively learn the hash code, image representation, and visual attributes more effectively. Experimental results on the popular ZSH datasets demonstrate that COMAE outperforms state-of-the-art hashing techniques, especially in scenarios with a larger number of unseen label classes.

Unsupervised hashing is important for indexing huge image or video collections without having expensive annotations available. Hashing aims to learn short binary codes for compact storage and efficient semantic retrieval. We propose an unsupervised deep hashing layer called Bi-half Net that maximizes entropy of the binary codes. Entropy is maximal when both possible values of the bit are uniformly (half-half) distributed. To maximize bit entropy, we do not add a term to the loss function as this is difficult to optimize and tune. Instead, we design a new parameter-free network layer to explicitly force continuous image features to approximate the optimal half-half bit distribution. This layer is shown to minimize a penalized term of the Wasserstein distance between the learned continuous image features and the optimal half-half bit distribution. Experimental results on the image datasets Flickr25k, Nus-wide, Cifar-10, Mscoco, Mnist and the video datasets Ucf-101 and Hmdb-51 show that our approach leads to compact codes and compares favorably to the current state-of-the-art.

This study introduces a novel BERT-LSH model that incorporates Locality Sensitive Hashing (LSH) to approximate the attention mechanism in the BERT architecture. We examine the computational efficiency and performance of this model compared to a standard baseline BERT model. Our findings reveal that BERT-LSH significantly reduces computational demand for the self-attention layer while unexpectedly outperforming the baseline model in pretraining and fine-tuning tasks. These results suggest that the LSH-based attention mechanism not only offers computational advantages but also may enhance the model’s ability to generalize from its training data. For more information, visit our GitHub repository: https://github.com/leo4life2/algoml-final

Recent years have witnessed wide application of hashing for large-scale image retrieval. However, most existing hashing methods are based on handcrafted features which might not be optimally compatible with the hashing procedure. Recently, deep hashing methods have been proposed to perform simultaneous feature learning and hash-code learning with deep neural networks, which have shown better performance than traditional hashing methods with hand-crafted features. Most of these deep hashing methods are supervised whose supervised information is given with triplet labels. For another common application scenario with pairwise labels, there have not existed methods for simultaneous feature learning and hash-code learning. In this paper, we propose a novel deep hashing method, called deep pairwise-supervised hashing (DPSH), to perform simultaneous feature learning and hashcode learning for applications with pairwise labels. Experiments on real datasets show that our DPSH method can outperform other methods to achieve the state-of-the-art performance in image retrieval applications.

We address the challenging large-scale content-based face image retrieval problem, intended as searching images based on the presence of specific subject, given one face image of him/her. To this end, one natural demand is a supervised binary code learning method. While the learned codes might be discriminating, people often have a further expectation that whether some semantic message (e.g., visual attributes) can be read from the human-incomprehensible codes. For this purpose, we propose a novel binary code learning framework by jointly encoding identity discriminability and a number of facial attributes into unified binary code. In this way, the learned binary codes can be applied to not only fine-grained face image retrieval, but also facial attributes prediction, which is the very innovation of this work, just like killing two birds with one stone. To evaluate the effectiveness of the proposed method, extensive experiments are conducted on a new purified large-scale web celebrity database, named CFW 60K, with abundant manual identity and attributes annotation, and experimental results exhibit the superiority of our method over state-of-the-art.

There has been considerable interest in random projections, an approximate algorithm for estimating distances between pairs of points in a high-dimensional vector space. Let A in Rn x D be our n points in D dimensions. The method multiplies A by a random matrix R in RD x k, reducing the D dimensions down to just k for speeding up the computation. R typically consists of entries of standard normal N(0,1). It is well known that random projections preserve pairwise distances (in the expectation). Achlioptas proposed sparse random projections by replacing the N(0,1) entries in R with entries in -1,0,1 with probabilities 1/6, 2/3, 1/6, achieving a threefold speedup in processing time.We recommend using R of entries in -1,0,1 with probabilities 1/2√D, 1-1√D, 1/2√D for achieving a significant √D-fold speedup, with little loss in accuracy.

Hashing has proven a valuable tool for large-scale information retrieval. Despite much success, existing hashing methods optimize over simple objectives such as the reconstruction error or graph Laplacian related loss functions, instead of the performance evaluation criteria of interest—multivariate performance measures such as the AUC and NDCG. Here we present a general framework (termed StructHash) that allows one to directly optimize multivariate performance measures. The resulting optimization problem can involve exponentially or infinitely many variables and constraints, which is more challenging than standard structured output learning. To solve the StructHash optimization problem, we use a combination of column generation and cutting-plane techniques. We demonstrate the generality of StructHash by applying it to ranking prediction and image retrieval, and show that it outperforms a few state-of-the-art hashing methods.

The MNIST database of handwritten digits, available from this page, has a training set of 60,000 examples, and a test set of 10,000 examples. It is a subset of a larger set available from NIST. The digits have been size-normalized and centered in a fixed-size image. It is a good database for people who want to try learning techniques and pattern recognition methods on real-world data while spending minimal efforts on preprocessing and formatting.

Recently, hashing based approximate nearest neighbors search has attracted much attention. Extensive centralized hashing algorithms have been proposed and achieved promising performance. However, due to the large scale of many applications, the data is often stored or even collected in a distributed manner. Learning hash functions by aggregating all the data into a fusion center is infeasible because of the prohibitively expensive communication and computation overhead. In this paper, we develop a novel hashing model to learn hash functions in a distributed setting. We cast a centralized hashing model as a set of subproblems with consensus constraints. We find these subproblems can be analytically solved in parallel on the distributed compute nodes. Since no training data is transmitted across the nodes in the learning process, the communication cost of our model is independent to the data size. Extensive experiments on several large scale datasets containing up to 100 million samples demonstrate the efficacy of our method.

Similarity-preserving hashing is a widely-used method for nearest neighbour search in large-scale image retrieval tasks. For most existing hashing methods, an image is first encoded as a vector of hand-engineering visual features, followed by another separate projection or quantization step that generates binary codes. However, such visual feature vectors may not be optimally compatible with the coding process, thus producing sub-optimal hashing codes. In this paper, we propose a deep architecture for supervised hashing, in which images are mapped into binary codes via carefully designed deep neural networks. The pipeline of the proposed deep architecture consists of three building blocks: 1) a sub-network with a stack of convolution layers to produce the effective intermediate image features; 2) a divide-and-encode module to divide the intermediate image features into multiple branches, each encoded into one hash bit; and 3) a triplet ranking loss designed to characterize that one image is more similar to the second image than to the third one. Extensive evaluations on several benchmark image datasets show that the proposed simultaneous feature learning and hash coding pipeline brings substantial improvements over other state-of-the-art supervised or unsupervised hashing methods.

The last few years have witnessed the rise of the big data era in which approximate nearest neighbor search is a fundamental problem in many applications, such as large-scale image retrieval. Recently, many research results have demonstrated that hashing can achieve promising performance due to its appealing storage and search efficiency. Since complex optimization problems for loss functions are difficult to solve, most hashing methods decompose the hash code learning problem into two steps: projection and quantization. In the quantization step, binary codes are widely used because ranking them by the Hamming distance is very efficient. However, the massive information loss produced by the quantization step should be reduced in applications where high search accuracy is required, such as in image retrieval. Since many two-step hashing methods produce uneven projected dimensions in the projection step, in this paper, we propose a novel dimension analysis-based quantization (DAQ) on two-step hashing methods for image retrieval. We first perform an importance analysis of the projected dimensions and select a subset of them that are more informative than others, and then we divide the selected projected dimensions into several regions with our quantizer. Every region is quantized with its corresponding codebook. Finally, the similarity between two hash codes is estimated by the Manhattan distance between their corresponding codebooks, which is also efficient. We conduct experiments on three public benchmarks containing up to one million descriptors and show that the proposed DAQ method consistently leads to significant accuracy improvements over state-of-the-art quantization methods.

Learning binary representations of instances and classes is a classical problem with several high potential applications. In modern settings, the compression of high-dimensional neural representations to low-dimensional binary codes is a challenging task and often require large bit-codes to be accurate. In this work, we propose a novel method for Learning Low-dimensional binary Codes (LLC) for instances as well as classes. Our method does not require any side-information, like annotated attributes or label meta-data, and learns extremely low-dimensional binary codes (~20 bits for ImageNet-1K). The learnt codes are super-efficient while still ensuring nearly optimal classification accuracy for ResNet50 on ImageNet-1K. We demonstrate that the learnt codes capture intrinsically important features in the data, by discovering an intuitive taxonomy over classes. We further quantitatively measure the quality of our codes by applying it to the efficient image retrieval as well as out-of-distribution (OOD) detection problems. For ImageNet-100 retrieval problem, our learnt binary codes outperform 16 bit HashNet using only 10 bits and also are as accurate as 10 dimensional real representations. Finally, our learnt binary codes can perform OOD detection, out-of-the-box, as accurately as a baseline that needs ~3000 samples to tune its threshold, while we require none.

Fast retrieval methods are critical for large-scale and data-driven vision applications. Recent work has explored ways to embed high-dimensional features or complex distance functions into a low-dimensional Hamming space where items can be efficiently searched. However, existing methods do not apply for high-dimensional kernelized data when the underlying feature embedding for the kernel is unknown. We show how to generalize locality-sensitive hashing to accommodate arbitrary kernel functions, making it possible to preserve the algorithm’s sub-linear time similarity search guarantees for a wide class of useful similarity functions. Since a number of successful image-based kernels have unknown or incomputable embeddings, this is especially valuable for image retrieval tasks. We validate our technique on several large-scale datasets, and show that it enables accurate and fast performance for example-based object classification, feature matching, and content-based retrieval.

Groups at MIT and NYU have collected a dataset of millions of tiny colour images from the web. It is, in principle, an excellent dataset for unsupervised training of deep generative models, but previous researchers who have tried this have found it difficult to learn a good set of filters from the images. We show how to train a multi-layer generative model that learns to extract meaningful features which resemble those found in the human visual cortex. Using a novel parallelization algorithm to distribute the work among multiple machines connected on a network, we show how training such a model can be done in reasonable time. A second problematic aspect of the tiny images dataset is that there are no reliable class labels which makes it hard to use for object recognition experiments. We created two sets of reliable labels. The CIFAR-10 set has 6000 examples of each of 10 classes and the CIFAR-100 set has 600 examples of each of 100 non-overlapping classes. Using these labels, we show that object recognition is significantly improved by pre-training a layer of features on a large set of unlabeled tiny images.

Fast retrieval methods are increasingly critical for many large-scale analysis tasks, and there have been several recent methods that attempt to learn hash functions for fast and accurate nearest neighbor searches. In this paper, we develop an algorithm for learning hash functions based on explicitly minimizing the reconstruction error between the original distances and the Hamming distances of the corresponding binary embeddings. We develop a scalable coordinate-descent algorithm for our proposed hashing objective that is able to efficiently learn hash functions in a variety of settings. Unlike existing methods such as semantic hashing and spectral hashing, our method is easily kernelized and does not require restrictive assumptions about the underlying distribution of the data. We present results over several domains to demonstrate that our method outperforms existing state-of-the-art techniques.

Hashing is used to learn binary-code representation for data with expectation of preserving the neighborhood structure in the original feature space. Due to its fast query speed and reduced storage cost, hashing has been widely used for efficient nearest neighbor search in a large variety of applications like text and image retrieval. Most existing hashing methods adopt Hamming distance to measure the similarity (neighborhood) between points in the hashcode space. However, one problem with Hamming distance is that it may destroy the neighborhood structure in the original feature space, which violates the essential goal of hashing. In this paper, Manhattan hashing (MH), which is based on Manhattan distance, is proposed to solve the problem of Hamming distance based hashing. The basic idea of MH is to encode each projected dimension with multiple bits of natural binary code (NBC), based on which the Manhattan distance between points in the hashcode space is calculated for nearest neighbor search. MH can effectively preserve the neighborhood structure in the data to achieve the goal of hashing. To the best of our knowledge, this is the first work to adopt Manhattan distance with NBC for hashing. Experiments on several largescale image data sets containing up to one million points show that our MH method can significantly outperform other state-of-the-art methods.

Most existing hashing methods adopt some projection functions to project the original data into several dimensions of real values, and then each of these projected dimensions is quantized into one bit (zero or one) by thresholding. Typically, the variances of different projected dimensions are different for existing projection functions such as principal component analysis (PCA). Using the same number of bits for different projected dimensions is unreasonable because larger-variance dimensions will carry more information. Although this viewpoint has been widely accepted by many researchers, it is still not verified by either theory or experiment because no methods have been proposed to find a projection with equal variances for different dimensions. In this paper, we propose a novel method, called isotropic hashing (IsoHash), to learn projection functions which can produce projected dimensions with isotropic variances (equal variances). Experimental results on real data sets show that IsoHash can outperform its counterpart with different variances for different dimensions, which verifies the viewpoint that projections with isotropic variances will be better than those with anisotropic variances.

Nearest neighbor-based similarity searching is a common task in chemistry, with notable use cases in drug discovery. Yet, some of the most commonly used approaches for this task still leverage a brute-force approach. In practice this can be computationally costly and overly time-consuming, due in part to the sheer size of modern chemical databases. Previous computational advancements for this task have generally relied on improvements to hardware or dataset-specific tricks that lack generalizability. Approaches that leverage lower-complexity searching algorithms remain relatively underexplored. However, many of these algorithms are approximate solutions and/or struggle with typical high-dimensional chemical embeddings. Here we evaluate whether a combination of low-dimensional chemical embeddings and a k-d tree data structure can achieve fast nearest neighbor queries while maintaining performance on standard chemical similarity search benchmarks. We examine different dimensionality reductions of standard chemical embeddings as well as a learned, structurally-aware embedding – SmallSA – for this task. With this framework, searches on over one billion chemicals execute in less than a second on a single CPU core, five orders of magnitude faster than the brute-force approach. We also demonstrate that SmallSA achieves competitive performance on chemical similarity benchmarks.

Hashing, which tries to learn similarity-preserving binary codes for data representation, has been widely used for efficient nearest neighbor search in massive databases due to its fast query speed and low storage cost. Because it is NP hard to directly compute the best binary codes for a given data set, mainstream hashing methods typically adopt a two-stage strategy. In the first stage, several projected dimensions of real values are generated. Then in the second stage, the real values will be quantized into binary codes by thresholding. Currently, most existing methods use one single bit to quantize each projected dimension. One problem with this single-bit quantization (SBQ) is that the threshold typically lies in the region of the highest point density and consequently a lot of neighboring points close to the threshold will be hashed to totally different bits, which is unexpected according to the principle of hashing. In this paper, we propose a novel quantization strategy, called double-bit quantization (DBQ), to solve the problem of SBQ. The basic idea of DBQ is to quantize each projected dimension into double bits with adaptively learned thresholds. Extensive experiments on two real data sets show that our DBQ strategy can signifi- cantly outperform traditional SBQ strategy for hashing.

Many applications in Multilingual and Multimodal Information Access involve searching large databases of high dimensional data objects with multiple (conditionally independent) views. In this work we consider the problem of learning hash functions for similarity search across the views for such applications. We propose a principled method for learning a hash function for each view given a set of multiview training data objects. The hash functions map similar objects to similar codes across the views thus enabling cross-view similarity search. We present results from an extensive empirical study of the proposed approach which demonstrate its effectiveness on Japanese language People Search and Multilingual People Search problems.

Redescription mining is a data analysis technique that has found applications in diverse fields. The most used redescription mining approaches involve two phases: finding matching pairs among data attributes and extending the pairs. This process is relatively efficient when the number of attributes remains limited and when the attributes are Boolean, but becomes almost intractable when the data consist of many numerical attributes. In this paper, we present new algorithms that perform the matching and extension orders of magnitude faster than the existing approaches. Our algorithms are based on locality-sensitive hashing with a tailored approach to handle the discretisation of numerical attributes as used in redescription mining.

Locality-sensitive hashing~[Indyk,Motwani’98] is a classical data structure for approximate nearest neighbor search. It allows, after a close to linear time preprocessing of the input dataset, to find an approximately nearest neighbor of any fixed query in sublinear time in the dataset size. The resulting data structure is randomized and succeeds with high probability for every fixed query. In many modern applications of nearest neighbor search the queries are chosen adaptively. In this paper, we study the robustness of the locality-sensitive hashing to adaptive queries in Hamming space. We present a simple adversary that can, under mild assumptions on the initial point set, provably find a query to the approximate near neighbor search data structure that the data structure fails on. Crucially, our adaptive algorithm finds the hard query exponentially faster than random sampling.

By leveraging semantic (label) information, supervised hashing has demonstrated better accuracy than unsupervised hashing in many real applications. Because the hashing-code learning problem is essentially a discrete optimization problem which is hard to solve, most existing supervised hashing methods try to solve a relaxed continuous optimization problem by dropping the discrete constraints. However, these methods typically suffer from poor performance due to the errors caused by the relaxation. Some other methods try to directly solve the discrete optimization problem. However, they are typically time-consuming and unscalable. In this paper, we propose a novel method, called column sampling based discrete supervised hashing (COSDISH), to directly learn the discrete hashing code from semantic information. COSDISH is an iterative method, in each iteration of which several columns are sampled from the semantic similarity matrix and then the hashing code is decomposed into two parts which can be alternately optimized in a discrete way. Theoretical analysis shows that the learning (optimization) algorithm of COSDISH has a constant-approximation bound in each step of the alternating optimization procedure. Empirical results on datasets with semantic labels illustrate that COSDISH can outperform the state-of-the-art methods in real applications like image retrieval.

Deep hashing enables computation and memory efficient image search through end-to-end learning of feature representations and binary codes. While linear scan over binary hash codes is more efficient than over the high-dimensional representations, its linear-time complexity is still unacceptable for very large databases. Hamming space retrieval enables constant-time search through hash lookups, where for each query, there is a Hamming ball centered at the query and the data points within the ball are returned as relevant. Since inside the Hamming ball implies retrievable while outside irretrievable, it is crucial to explicitly characterize the Hamming ball. The main idea of this work is to directly embody the Hamming radius into the loss functions, leading to Maximum-Margin Hamming Hashing (MMHH), a new model specifically optimized for Hamming space retrieval. We introduce a max-margin t-distribution loss, where the t-distribution concentrates more similar data points to be within the Hamming ball, and the margin characterizes the Hamming radius such that less penalization is applied to similar data points within the Hamming ball. The loss function also introduces robustness to data noise, where the similarity supervision may be inaccurate in practical problems. The model is trained end-to-end using a new semi-batch optimization algorithm tailored to extremely imbalanced data. Our method yields state-of-the-art results on four datasets and shows superior performance on noisy data.

Following the success of hashing methods for multidimensional indexing, more and more works are interested in embedding visual feature space in compact hash codes. Such approaches are not an alternative to using index structures but a complementary way to reduce both the memory usage and the distance computation cost. Several data dependent hash functions have notably been proposed to closely fit data distribution and provide better selectivity than usual random projections such as LSH. However, improvements occur only for relatively small hash code sizes up to 64 or 128 bits. As discussed in the paper, this is mainly due to the lack of independence between the produced hash functions. We introduce a new hash function family that attempts to solve this issue in any kernel space. Rather than boosting the collision probability of close points, our method focus on data scattering. By training purely random splits of the data, regardless the closeness of the training samples, it is indeed possible to generate consistently more independent hash functions. On the other side, the use of large margin classifiers allows to maintain good generalization performances. Experiments show that our new Random Maximum Margin Hashing scheme (RMMH) outperforms four state-of-the-art hashing methods, notably in kernel spaces.

Deep hashing methods have been proved to be effective and efficient for large-scale Web media search. The success of these data-driven methods largely depends on collecting sufficient labeled data, which is usually a crucial limitation in practical cases. The current solutions to this issue utilize Generative Adversarial Network (GAN) to augment data in semi-supervised learning. However, existing GAN-based methods treat image generations and hashing learning as two isolated processes, leading to generation ineffectiveness. Besides, most works fail to exploit the semantic information in unlabeled data. In this paper, we propose a novel Semi-supervised Self-pace Adversarial Hashing method, named SSAH to solve the above problems in a unified framework. The SSAH method consists of an adversarial network (A-Net) and a hashing network (H-Net). To improve the quality of generative images, first, the A-Net learns hard samples with multi-scale occlusions and multi-angle rotated deformations which compete against the learning of accurate hashing codes. Second, we design a novel self-paced hard generation policy to gradually increase the hashing difficulty of generated samples. To make use of the semantic information in unlabeled ones, we propose a semi-supervised consistent loss. The experimental results show that our method can significantly improve state-of-the-art models on both the widely-used hashing datasets and fine-grained datasets.

In recent years, hashing methods have been proved to be effective and efficient for the large-scale Web media search. However, the existing general hashing methods have limited discriminative power for describing fine-grained objects that share similar overall appearance but have subtle difference. To solve this problem, we for the first time introduce the attention mechanism to the learning of fine-grained hashing codes. Specifically, we propose a novel deep hashing model, named deep saliency hashing (DSaH), which automatically mines salient regions and learns semanticpreserving hashing codes simultaneously. DSaH is a twostep end-to-end model consisting of an attention network and a hashing network. Our loss function contains three basic components, including the semantic loss, the saliency loss, and the quantization loss. As the core of DSaH, the saliency loss guides the attention network to mine discriminative regions from pairs of images. We conduct extensive experiments on both fine-grained and general retrieval datasets for performance evaluation. Experimental results on fine grained dataset, including Oxford Flowers-17, Stanford Dogs-120 and CUB Bird demonstrate that our DSaH performs the best for fine-grained retrieval task and beats strongest competitor (DTQ) by approximately 10% on both Stanford Dogs-120 and CUB Bird. DSaH is also comparable to several state-of-the-art hashing methods on general datasets, including CIFAR-10 and NUS-WIDE.

In recent years, supervised hashing has been validated to greatly boost the performance of image retrieval. However, the label-hungry property requires massive label collection, making it intractable in practical scenarios. To liberate the model training procedure from laborious manual annotations, some unsupervised methods are proposed. However, the following two factors make unsupervised algorithms inferior to their supervised counterparts: (1) Without manually-defined labels, it is difficult to capture the semantic information across data, which is of crucial importance to guide robust binary code learning. (2) The widely adopted relaxation on binary constraints results in quantization error accumulation in the optimization procedure. To address the above-mentioned problems, in this paper, we propose a novel Unsupervised Discrete Hashing method (UDH). Specifically, to capture the semantic information, we propose a balanced graph-based semantic loss which explores the affinity priors in the original feature space. Then, we propose a novel self-supervised loss, termed orthogonal consistent loss, which can leverage semantic loss of instance and impose independence of codes. Moreover, by integrating the discrete optimization into the proposed unsupervised framework, the binary constraints are consistently preserved, alleviating the influence of quantization errors. Extensive experiments demonstrate that UDH outperforms state-of-the-art unsupervised methods for image retrieval.

Recently, hashing techniques have been widely applied to solve the approximate nearest neighbors search problem in many vision applications. Generally, these hashing approaches generate 2^c buckets, where c is the length of the hash code. A good hashing method should satisfy the following two requirements: 1) mapping the nearby data points into the same bucket or nearby (measured by the Hamming distance) buckets. 2) all the data points are evenly distributed among all the buckets. In this paper, we propose a novel algorithm named Complementary Projection Hashing (CPH) to find the optimal hashing functions which explicitly considers the above two requirements. Specifically, CPH aims at sequentially finding a series of hyperplanes (hashing functions) which cross the sparse region of the data. At the same time, the data points are evenly distributed in the hypercubes generated by these hyperplanes. The experiments comparing with the state-of-the-art hashing methods demonstrate the effectiveness of the proposed method.

Due to its low storage cost and fast query speed, crossmodal hashing (CMH) has been widely used for similarity search in multimedia retrieval applications. However, most existing CMH methods are based on hand-crafted features which might not be optimally compatible with the hash-code learning procedure. As a result, existing CMH methods with hand-crafted features may not achieve satisfactory performance. In this paper, we propose a novel CMH method, called deep cross-modal hashing (DCMH), by integrating feature learning and hash-code learning into the same framework. DCMH is an end-to-end learning framework with deep neural networks, one for each modality, to perform feature learning from scratch. Experiments on three real datasets with image-text modalities show that DCMH can outperform other baselines to achieve the state-of-the-art performance in cross-modal retrieval applications.

Hashing has been widely used for approximate nearest neighbor (ANN) search in big data applications because of its low storage cost and fast retrieval speed. The goal of hashing is to map the data points from the original space into a binary-code space where the similarity (neighborhood structure) in the original space is preserved. By directly exploiting the similarity to guide the hashing code learning procedure, graph hashing has attracted much attention. However, most existing graph hashing methods cannot achieve satisfactory performance in real applications due to the high complexity for graph modeling. In this paper, we propose a novel method, called scalable graph hashing with feature transformation (SGH), for large-scale graph hashing. Through feature transformation, we can effectively approximate the whole graph without explicitly computing the similarity graph matrix, based on which a sequential learning method is proposed to learn the hash functions in a bit-wise manner. Experiments on two datasets with one million data points show that our SGH method can outperform the state-of-the-art methods in terms of both accuracy and scalability.

Hashing has been widely researched to solve the large-scale approximate nearest neighbor search problem owing to its time and storage superiority. In recent years, a number of online hashing methods have emerged, which can update the hash functions to adapt to the new stream data and realize dynamic retrieval. However, existing online hashing methods are required to update the whole database with the latest hash functions when a query arrives, which leads to low retrieval efficiency with the continuous increase of the stream data. On the other hand, these methods ignore the supervision relationship among the examples, especially in the multi-label case. In this paper, we propose a novel Fast Online Hashing (FOH) method which only updates the binary codes of a small part of the database. To be specific, we first build a query pool in which the nearest neighbors of each central point are recorded. When a new query arrives, only the binary codes of the corresponding potential neighbors are updated. In addition, we create a similarity matrix which takes the multi-label supervision information into account and bring in the multi-label projection loss to further preserve the similarity among the multi-label data. The experimental results on two common benchmarks show that the proposed FOH can achieve dramatic superiority on query time up to 6.28 seconds less than state-of-the-art baselines with competitive retrieval accuracy.

We propose an approximate nearest neighbor search method based on quantization. It uses, in particular, product quantizer to produce short codes and corresponding distance estimators approximating the Euclidean distance between the orginal vectors. The method is advantageously used in an asymmetric manner, by computing the distance between a vector and code, unlike competing techniques such as spectral hashing that only compare codes. Our approach approximates the Euclidean distance based on memory efficient codes and, thus, permits efficient nearest neighbor search. Experiments performed on SIFT and GIST image descriptors show excellent search accuracy. The method is shown to outperform two state-of-the-art approaches of the literature. Timings measured when searching a vector set of 2 billion vectors are shown to be excellent given the high accuracy of the method.

``Learning to hash’’ is a practical solution for efficient retrieval, offering fast search speed and low storage cost. It is widely applied in various applications, such as image-text cross-modal search. In this paper, we explore the potential of enhancing the performance of learning to hash with the proliferation of powerful large pre-trained models, such as Vision-Language Pre-training (VLP) models. We introduce a novel method named Distillation for Cross-Modal Quantization (DCMQ), which leverages the rich semantic knowledge of VLP models to improve hash representation learning. Specifically, we use the VLP as a teacher' to distill knowledge into a student’ hashing model equipped with codebooks. This process involves the replacement of supervised labels, which are composed of multi-hot vectors and lack semantics, with the rich semantics of VLP. In the end, we apply a transformation termed Normalization with Paired Consistency (NPC) to achieve a discriminative target for distillation. Further, we introduce a new quantization method, Product Quantization with Gumbel (PQG) that promotes balanced codebook learning, thereby improving the retrieval performance. Extensive benchmark testing demonstrates that DCMQ consistently outperforms existing supervised cross-modal hashing approaches, showcasing its significant potential.

We propose a method to efficiently index into a large database of examples according to a learned metric. Given a collection of examples, we learn a Mahalanobis distance using an information-theoretic metric learning technique that adapts prior knowledge about pairwise distances to incorporate similarity and dissimilarity constraints. To enable sub-linear time similarity search under the learned metric, we show how to encode a learned Mahalanobis parameterization into randomized locality-sensitive hash functions. We further formulate an indirect solution that enables metric learning and hashing for sparse input vector spaces whose high dimensionality make it infeasible to learn an explicit weighting over the feature dimensions. We demonstrate the approach applied to systems and image datasets, and show that our learned metrics improve accuracy relative to commonly-used metric baselines, while our hashing construction permits effi- cient indexing with a learned distance and very large databases.

We consider the problem of retrieving the database points nearest to a given hyperplane query without exhaustively scanning the database. We propose two hashing-based solutions. Our first approach maps the data to two-bit binary keys that are locality-sensitive for the angle between the hyperplane normal and a database point. Our second approach embeds the data into a vector space where the Euclidean norm reflects the desired distance between the original points and hyperplane query. Both use hashing to retrieve near points in sub-linear time. Our first method’s preprocessing stage is more efficient, while the second has stronger accuracy guarantees. We apply both to pool-based active learning: taking the current hyperplane classifier as a query, our algorithm identifies those points (approximately) satisfying the well-known minimal distance-to-hyperplane selection criterion. We empirically demonstrate our methods’ tradeoffs, and show that they make it practical to perform active selection with millions of unlabeled points.

We give new data-dependent locality sensitive hashing schemes (LSH) for the Earth Mover’s Distance (\(\mathsf{EMD}\)), and as a result, improve the best approximation for nearest neighbor search under \(\mathsf{EMD}\) by a quadratic factor. Here, the metric \(\mathsf{EMD}_s(\mathbb{R}^d,\ell_p)\) consists of sets of \(s\) vectors in \(\mathbb{R}^d\), and for any two sets \(x,y\) of \(s\) vectors the distance \(\mathsf{EMD}(x,y)\) is the minimum cost of a perfect matching between \(x,y\), where the cost of matching two vectors is their \(\ell_p\) distance. Previously, Andoni, Indyk, and Krauthgamer gave a (data-independent) locality-sensitive hashing scheme for \(\mathsf{EMD}_s(\mathbb{R}^d,\ell_p)\) when \(p \in [1,2]\) with approximation \(O(log^2 s)\). By being data-dependent, we improve the approximation to \(\tilde{O}(log s)\). Our main technical contribution is to show that for any distribution \(\mu\) supported on the metric \(\mathsf{EMD}_s(\mathbb{R}^d, \ell_p)\), there exists a data-dependent LSH for dense regions of \(\mu\) which achieves approximation \(\tilde{O}(log s)\), and that the data-independent LSH actually achieves a \(\tilde{O}(log s)\)-approximation outside of those dense regions. Finally, we show how to “glue” together these two hashing schemes without any additional loss in the approximation. Beyond nearest neighbor search, our data-dependent LSH also gives optimal (distributional) sketches for the Earth Mover’s Distance. By known sketching lower bounds, this implies that our LSH is optimal (up to \(\mathrm{poly}(log log s)\) factors) among those that collide close points with constant probability.

Spatial optimization is often overlooked in many computer vision tasks. Filters should be able to recognize the features of an object regardless of where it is in the image. Similarity search is a crucial task where spatial features decide an important output. The capacity of convolution to capture visual patterns across various locations is limited. In contrast to convolution, the involution kernel is dynamically created at each pixel based on the pixel value and parameters that have been learned. This study demonstrates that utilizing a single layer of involution feature extractor alongside a compact convolution model significantly enhances the performance of similarity search. Additionally, we improve predictions by using the GELU activation function rather than the ReLU. The negligible amount of weight parameters in involution with a compact model with better performance makes the model very useful in real-world implementations. Our proposed model is below 1 megabyte in size. We have experimented with our proposed methodology and other models on CIFAR-10, FashionMNIST, and MNIST datasets. Our proposed method outperforms across all three datasets.

Previous efforts in hashing intend to preserve data variance or pairwise affinity, but neither is adequate in capturing the manifold structures hidden in most visual data. In this paper, we tackle this problem by reconstructing the locally linear structures of manifolds in the binary Hamming space, which can be learned by locality-sensitive sparse coding. We cast the problem as a joint minimization of reconstruction error and quantization loss, and show that, despite its NP-hardness, a local optimum can be obtained efficiently via alternative optimization. Our method distinguishes itself from existing methods in its remarkable ability to extract the nearest neighbors of the query from the same manifold, instead of from the ambient space. On extensive experiments on various image benchmarks, our results improve previous state-of-the-art by 28-74% typically, and 627% on the Yale face data.

BIGANN consists of SIFT descriptors applied to images from extracted from a large image dataset.

Recent years have witnessed the success of learning to hash in fast large-scale image retrieval. As deep learning has shown its superior performance on many computer vision applications, recent designs of learning-based hashing models have been moving from shallow ones to deep architectures. However, based on our analysis, we find that gradient descent based algorithms used in deep hashing models would potentially cause hash codes of a pair of training instances to be updated towards the directions of each other simultaneously during optimization. In the worst case, the paired hash codes switch their directions after update, and consequently, their corresponding distance in the Hamming space remain unchanged. This makes the overall learning process highly inefficient. To address this issue, we propose a new deep hashing model integrated with a novel gradient attention mechanism. Extensive experimental results on three benchmark datasets show that our proposed algorithm is able to accelerate the learning process and obtain competitive retrieval performance compared with state-of-the-art deep hashing models.

Cross-modal hashing, due to its low storage cost and high query speed, has been successfully used for similarity search in multimedia retrieval applications. It projects high-dimensional data into a shared isomorphic Hamming space with similar binary codes for semantically-similar data. In some applications, all modalities may not be obtained or trained simultaneously for some reasons, such as privacy, secret, storage limitation, and computational resource limitation. However, most existing cross-modal hashing methods need all modalities to jointly learn the common Hamming space, thus hindering them from handling these problems. In this paper, we propose a novel approach called Separated Variational Hashing Networks (SVHNs) to overcome the above challenge. Firstly, it adopts a label network (LabNet) to exploit available and nonspecific label annotations to learn a latent common Hamming space by projecting each semantic label into a common binary representation. Then, each modality-specific network can separately map the samples of the corresponding modality into their binary semantic codes learned by LabNet. We achieve it by conducting variational inference to match the aggregated posterior of the hashing code of LabNet with an arbitrary prior distribution. The effectiveness and efficiency of our SVHNs are verified by extensive experiments carried out on four widely-used multimedia databases, in comparison with 11 state-of-the-art approaches.

Vector quantization is a fundamental operation for data compression and vector search. To obtain high accuracy, multi-codebook methods represent each vector using codewords across several codebooks. Residual quantization (RQ) is one such method, which iteratively quantizes the error of the previous step. While the error distribution is dependent on previously-selected codewords, this dependency is not accounted for in conventional RQ as it uses a fixed codebook per quantization step. In this paper, we propose QINCo, a neural RQ variant that constructs specialized codebooks per step that depend on the approximation of the vector from previous steps. Experiments show that QINCo outperforms state-of-the-art methods by a large margin on several datasets and code sizes. For example, QINCo achieves better nearest-neighbor search accuracy using 12-byte codes than the state-of-the-art UNQ using 16 bytes on the BigANN1M and Deep1M datasets.

Recent watermarked generation algorithms inject detectable signatures during language generation to facilitate post-hoc detection. While token-level watermarks are vulnerable to paraphrase attacks, SemStamp (Hou et al., 2023) applies watermark on the semantic representation of sentences and demonstrates promising robustness. SemStamp employs locality-sensitive hashing (LSH) to partition the semantic space with arbitrary hyperplanes, which results in a suboptimal tradeoff between robustness and speed. We propose k-SemStamp, a simple yet effective enhancement of SemStamp, utilizing k-means clustering as an alternative of LSH to partition the embedding space with awareness of inherent semantic structure. Experimental results indicate that k-SemStamp saliently improves its robustness and sampling efficiency while preserving the generation quality, advancing a more effective tool for machine-generated text detection.

In recent years, cross-modal hashing (CMH) has attracted increasing attentions, mainly because its potential ability of mapping contents from different modalities, especially in vision and language, into the same space, so that it becomes efficient in cross-modal data retrieval. There are two main frameworks for CMH, differing from each other in whether semantic supervision is required. Compared to the unsupervised methods, the supervised methods often enjoy more accurate results, but require much heavier labors in data annotation. In this paper, we propose a novel approach that enables guiding a supervised method using outputs produced by an unsupervised method. Specifically, we make use of teacher-student optimization for propagating knowledge. Experiments are performed on two popular CMH benchmarks, i.e., the MIRFlickr and NUS-WIDE datasets. Our approach outperforms all existing unsupervised methods by a large margin

In most well known image retrieval test sets, the imagery typically cannot be freely distributed or is not representative of a large community of users. In this paper we present a collection for the MIR community comprising 25000 images from the Flickr website which are redistributable for research purposes and represent a real community of users both in the image content and image tags. We have extracted the tags and EXIF image metadata, and also make all of these publicly available. In addition we discuss several challenges for benchmarking retrieval and classification methods.

In the era of big data, methods for improving memory and computational efficiency have become crucial for successful deployment of technologies. Hashing is one of the most effective approaches to deal with computational limitations that come with big data. One natural way for formulating this problem is spectral hashing that directly incorporates affinity to learn binary codes. However, due to binary constraints, the optimization becomes intractable. To mitigate this challenge, different relaxation approaches have been proposed to reduce the computational load of obtaining binary codes and still attain a good solution. The problem with all existing relaxation methods is resorting to one or more additional auxiliary variables to attain high quality binary codes while relaxing the problem. The existence of auxiliary variables leads to coordinate descent approach which increases the computational complexity. We argue that introducing these variables is unnecessary. To this end, we propose a novel relaxed formulation for spectral hashing that adds no additional variables to the problem. Furthermore, instead of solving the problem in original space where number of variables is equal to the data points, we solve the problem in a much smaller space and retrieve the binary codes from this solution. This trick reduces both the memory and computational complexity at the same time. We apply two optimization techniques, namely projected gradient and optimization on manifold, to obtain the solution. Using comprehensive experiments on four public datasets, we show that the proposed efficient spectral hashing (ESH) algorithm achieves highly competitive retrieval performance compared with state of the art at low complexity.

Many binary code encoding schemes based on hashing have been actively studied recently, since they can provide efficient similarity search, especially nearest neighbor search, and compact data representations suitable for handling large scale image databases in many computer vision problems. Existing hashing techniques encode highdimensional data points by using hyperplane-based hashing functions. In this paper we propose a novel hyperspherebased hashing function, spherical hashing, to map more spatially coherent data points into a binary code compared to hyperplane-based hashing functions. Furthermore, we propose a new binary code distance function, spherical Hamming distance, that is tailored to our hyperspherebased binary coding scheme, and design an efficient iterative optimization process to achieve balanced partitioning of data points for each hash function and independence between hashing functions. Our extensive experiments show that our spherical hashing technique significantly outperforms six state-of-the-art hashing techniques based on hyperplanes across various image benchmarks of sizes ranging from one to 75 million of GIST descriptors. The performance gains are consistent and large, up to 100% improvements. The excellent results confirm the unique merits of the proposed idea in using hyperspheres to encode proximity regions in high-dimensional spaces. Finally, our method is intuitive and easy to implement.

On the GPU, hash table operation speed is determined in large part by cache line efficiency, and state-of-the-art hashing schemes thus divide tables into cache line-sized buckets. This raises the question whether performance can be further improved by increasing the number of entries that fit in such buckets. Known compact hashing techniques have not yet been adapted to the massively parallel setting, nor have they been evaluated on the GPU. We consider a compact version of bucketed cuckoo hashing, and a version of compact iceberg hashing suitable for the GPU. We discuss the tables from a theoretical perspective, and provide an open source implementation of both schemes in CUDA for comparative benchmarking. In terms of performance, the state-of-the-art cuckoo hashing benefits from compactness on lookups and insertions (most experiments show at least 10-20% increase in throughput), and the iceberg table benefits significantly, to the point of being comparable to compact cuckoo hashing–while supporting performant dynamic operation.

An effective unsupervised hashing algorithm leads to compact binary codes preserving the neighborhood structure of data as much as possible. One of the most established schemes for unsupervised hashing is to reduce the dimensionality of data and then find a rigid (neighbourhood-preserving) transformation that reduces the quantization error. Although employing rigid transformations is effective, we may not reduce quantization loss to the ultimate limits. As well, reducing dimensionality and quantization loss in two separate steps seems to be sub-optimal. Motivated by these shortcomings, we propose to employ both rigid and non-rigid transformations to reduce quantization error and dimensionality simultaneously. We relax the orthogonality constraint on the projection in a PCA-formulation and regularize this by a quantization term. We show that both the non-rigid projection matrix and rotation matrix contribute towards minimizing quantization loss but in different ways. A scalable nested coordinate descent approach is proposed to optimize this mixed-integer optimization problem. We evaluate the proposed method on five public benchmark datasets providing almost half a million images. Comparative results indicate that the proposed method mostly outperforms state-of-art linear methods and competes with end-to-end deep solutions.

A minimal perfect hash function (MPHF) maps a set of n keys to {1, …, n} without collisions. Such functions find widespread application e.g. in bioinformatics and databases. In this paper we revisit PTHash - a construction technique particularly designed for fast queries. PTHash distributes the input keys into small buckets and, for each bucket, it searches for a hash function seed that places its keys in the output domain without collisions. The collection of all seeds is then stored in a compressed way. Since the first buckets are easier to place, buckets are considered in non-increasing order of size. Additionally, PTHash heuristically produces an imbalanced distribution of bucket sizes by distributing 60% of the keys into 30% of the buckets. Our main contribution is to characterize, up to lower order terms, an optimal distribution of expected bucket sizes. We arrive at a simple, closed form solution which improves construction throughput for space efficient configurations in practice. Our second contribution is a novel encoding scheme for the seeds. We split the keys into partitions. Within each partition, we run the bucket distribution and search step. We then store the seeds in an interleaved way by consecutively placing the seeds for the i-th buckets from all partitions. The seeds for the i-th bucket of each partition follow the same statistical distribution. This allows us to tune a compressor for each bucket. Hence, we call our technique PHOBIC - Perfect Hashing with Optimized Bucket sizes and Interleaved Coding. Compared to PTHash, PHOBIC is 0.17 bits/key more space efficient for same query time and construction throughput. We also contribute a GPU implementation to further accelerate MPHF construction. For a configuration with fast queries, PHOBIC-GPU can construct a perfect hash function at 2.17 bits/key in 28 ns per key, which can be queried in 37 ns on the CPU.

A deep hashing model typically has two main learning objectives: to make the learned binary hash codes discriminative and to minimize a quantization error. With further constraints such as bit balance and code orthogonality, it is not uncommon for existing models to employ a large number (>4) of losses. This leads to difficulties in model training and subsequently impedes their effectiveness. In this work, we propose a novel deep hashing model with only a single learning objective. Specifically, we show that maximizing the cosine similarity between the continuous codes and their corresponding binary orthogonal codes can ensure both hash code discriminativeness and quantization error minimization. Further, with this learning objective, code balancing can be achieved by simply using a Batch Normalization (BN) layer and multi-label classification is also straightforward with label smoothing. The result is an one-loss deep hashing model that removes all the hassles of tuning the weights of various losses. Importantly, extensive experiments show that our model is highly effective, outperforming the state-of-the-art multi-loss hashing models on three large-scale instance retrieval benchmarks, often by significant margins.

Hashing, or learning binary embeddings of data, is frequently used in nearest neighbor retrieval. In this paper, we develop learning to rank formulations for hashing, aimed at directly optimizing ranking-based evaluation metrics such as Average Precision (AP) and Normalized Discounted Cumulative Gain (NDCG). We first observe that the integer-valued Hamming distance often leads to tied rankings, and propose to use tie-aware versions of AP and NDCG to evaluate hashing for retrieval. Then, to optimize tie-aware ranking metrics, we derive their continuous relaxations, and perform gradient-based optimization with deep neural networks. Our results establish the new state-of-the-art for image retrieval by Hamming ranking in common benchmarks.

Unsupervised semantic hashing has emerged as an indispensable technique for fast image search, which aims to convert images into binary hash codes without relying on labels. Recent advancements in the field demonstrate that employing large-scale backbones (e.g., ViT) in unsupervised semantic hashing models can yield substantial improvements. However, the inference delay has become increasingly difficult to overlook. Knowledge distillation provides a means for practical model compression to alleviate this delay. Nevertheless, the prevailing knowledge distillation approaches are not explicitly designed for semantic hashing. They ignore the unique search paradigm of semantic hashing, the inherent necessities of the distillation process, and the property of hash codes. In this paper, we propose an innovative Bit-mask Robust Contrastive knowledge Distillation (BRCD) method, specifically devised for the distillation of semantic hashing models. To ensure the effectiveness of two kinds of search paradigms in the context of semantic hashing, BRCD first aligns the semantic spaces between the teacher and student models through a contrastive knowledge distillation objective. Additionally, to eliminate noisy augmentations and ensure robust optimization, a cluster-based method within the knowledge distillation process is introduced. Furthermore, through a bit-level analysis, we uncover the presence of redundancy bits resulting from the bit independence property. To mitigate these effects, we introduce a bit mask mechanism in our knowledge distillation objective. Finally, extensive experiments not only showcase the noteworthy performance of our BRCD method in comparison to other knowledge distillation methods but also substantiate the generality of our methods across diverse semantic hashing models and backbones. The code for BRCD is available at https://github.com/hly1998/BRCD.

Deep image hashing aims to map input images into simple binary hash codes via deep neural networks and thus enable effective large-scale image retrieval. Recently, hybrid networks that combine convolution and Transformer have achieved superior performance on various computer tasks and have attracted extensive attention from researchers. Nevertheless, the potential benefits of such hybrid networks in image retrieval still need to be verified. To this end, we propose a hybrid convolutional and self-attention deep hashing method known as HybridHash. Specifically, we propose a backbone network with stage-wise architecture in which the block aggregation function is introduced to achieve the effect of local self-attention and reduce the computational complexity. The interaction module has been elaborately designed to promote the communication of information between image blocks and to enhance the visual representations. We have conducted comprehensive experiments on three widely used datasets: CIFAR-10, NUS-WIDE and IMAGENET. The experimental results demonstrate that the method proposed in this paper has superior performance with respect to state-of-the-art deep hashing methods. Source code is available https://github.com/shuaichaochao/HybridHash.

Hashing based approximate nearest neighbor search embeds high dimensional data to compact binary codes, which enables efficient similarity search and storage. However, the non-isometry sign(·) function makes it hard to project the nearest neighbors in continuous data space into the closest codewords in discrete Hamming space. In this work, we revisit the sign(·) function from the perspective of space partitioning. In specific, we bridge the gap between k-nearest neighbors and binary hashing codes with Shannon entropy. We further propose a novel K-Nearest Neighbors Hashing (KNNH) method to learn binary representations from KNN within the subspaces generated by sign(·). Theoretical and experimental results show that the KNN relation is of central importance to neighbor preserving embeddings, and the proposed method outperforms the state-of-the-arts on benchmark datasets.

Orthogonal arrays are a type of combinatorial design that were developed in the 1940s in the design of statistical experiments. In 1947, Rao proved a lower bound on the size of any orthogonal array, and raised the problem of constructing arrays of minimum size. Kuperberg, Lovett and Peled (2017) gave a non-constructive existence proof of orthogonal arrays whose size is near-optimal (i.e., within a polynomial of Rao’s lower bound), leaving open the question of an algorithmic construction. We give the first explicit, deterministic, algorithmic construction of orthogonal arrays achieving near-optimal size for all parameters. Our construction uses algebraic geometry codes. In pseudorandomness, the notions of \(t\)-independent generators or \(t\)-independent hash functions are equivalent to orthogonal arrays. Classical constructions of \(t\)-independent hash functions are known when the size of the codomain is a prime power, but very few constructions are known for an arbitrary codomain. Our construction yields algorithmically efficient \(t\)-independent hash functions for arbitrary domain and codomain.

Content-aware recommendation approaches are essential for providing meaningful recommendations for new (i.e., cold-start) items in a recommender system. We present a content-aware neural hashing-based collaborative filtering approach (NeuHash-CF), which generates binary hash codes for users and items, such that the highly efficient Hamming distance can be used for estimating user-item relevance. NeuHash-CF is modelled as an autoencoder architecture, consisting of two joint hashing components for generating user and item hash codes. Inspired from semantic hashing, the item hashing component generates a hash code directly from an item’s content information (i.e., it generates cold-start and seen item hash codes in the same manner). This contrasts existing state-of-the-art models, which treat the two item cases separately. The user hash codes are generated directly based on user id, through learning a user embedding matrix. We show experimentally that NeuHash-CF significantly outperforms state-of-the-art baselines by up to 12% NDCG and 13% MRR in cold-start recommendation settings, and up to 4% in both NDCG and MRR in standard settings where all items are present while training. Our approach uses 2-4x shorter hash codes, while obtaining the same or better performance compared to the state of the art, thus consequently also enabling a notable storage reduction.

Semantic Hashing is a popular family of methods for efficient similarity search in large-scale datasets. In Semantic Hashing, documents are encoded as short binary vectors (i.e., hash codes), such that semantic similarity can be efficiently computed using the Hamming distance. Recent state-of-the-art approaches have utilized weak supervision to train better performing hashing models. Inspired by this, we present Semantic Hashing with Pairwise Reconstruction (PairRec), which is a discrete variational autoencoder based hashing model. PairRec first encodes weakly supervised training pairs (a query document and a semantically similar document) into two hash codes, and then learns to reconstruct the same query document from both of these hash codes (i.e., pairwise reconstruction). This pairwise reconstruction enables our model to encode local neighbourhood structures within the hash code directly through the decoder. We experimentally compare PairRec to traditional and state-of-the-art approaches, and obtain significant performance improvements in the task of document similarity search.

Microsoft Turing-ANNS-1B is a new dataset being released by the Microsoft Turing team for this competition. It consists of Bing queries encoded by Turing AGI v5 that trains Transformers to capture similarity of intent in web search queries. An early version of the RNN-based AGI Encoder is described in a SIGIR’19 paper and a blogpost.

Modern financial institutions rely on data for many operations, including a need to drive efficiency, enhance services and prevent financial crime. Data sharing across an organisation or between institutions can facilitate rapid, evidence-based decision making, including identifying money laundering and fraud. However, data privacy regulations impose restrictions on data sharing. Privacy-enhancing technologies are being increasingly employed to allow organisations to derive shared intelligence while ensuring regulatory compliance. This paper examines the case in which regulatory restrictions mean a party cannot share data on accounts of interest with another (internal or external) party to identify people that hold an account in each dataset. We observe that the names of account holders may be recorded differently in each data set. We introduce a novel privacy-preserving approach for fuzzy name matching across institutions, employing fully homomorphic encryption with locality-sensitive hashing. The efficiency of the approach is enhanced using a clustering mechanism. The practicality and effectiveness of the proposed approach are evaluated using different datasets. Experimental results demonstrate it takes around 100 and 1000 seconds to search 1000 names from 10k and 100k names, respectively. Moreover, the proposed approach exhibits significant improvement in reducing communication overhead by 30-300 times, using clustering.

Algorithms to rapidly search massive image or video collections are critical for many vision applications, including visual search, content-based retrieval, and non-parametric models for object recognition. Recent work shows that learned binary projections are a powerful way to index large collections according to their content. The basic idea is to formulate the projections so as to approximately preserve a given similarity function of interest. Having done so, one can then search the data efficiently using hash tables, or by exploring the Hamming ball volume around a novel query. Both enable sub-linear time retrieval with respect to the database size. Further, depending on the design of the projections, in some cases it is possible to bound the number of database examples that must be searched in order to achieve a given level of accuracy.

This chapter overviews data structures for fast search with binary codes, and then describes several supervised and unsupervised strategies for generating the codes. In particular, we review supervised methods that integrate metric learning, boosting, and neural networks into the hash key construction, and unsupervised methods based on spectral analysis or kernelized random projections that compute affinity-preserving binary codes.Whether learning from explicit semantic supervision or exploiting the structure among unlabeled data, these methods make scalable retrieval possible for a variety of robust visual similarity measures.We focus on defining the algorithms, and illustrate the main points with results using millions of images.

We present Lagrangian Hashing, a representation for neural fields combining the characteristics of fast training NeRF methods that rely on Eulerian grids (i.e.~InstantNGP), with those that employ points equipped with features as a way to represent information (e.g. 3D Gaussian Splatting or PointNeRF). We achieve this by incorporating a point-based representation into the high-resolution layers of the hierarchical hash tables of an InstantNGP representation. As our points are equipped with a field of influence, our representation can be interpreted as a mixture of Gaussians stored within the hash table. We propose a loss that encourages the movement of our Gaussians towards regions that require more representation budget to be sufficiently well represented. Our main finding is that our representation allows the reconstruction of signals using a more compact representation without compromising quality.

Recent advances in visual recognition indicate that to achieve good retrieval and classification accuracy on largescale datasets like ImageNet, extremely high-dimensional visual descriptors, e.g., Fisher Vectors, are needed. We present a novel method for converting such descriptors to compact similarity-preserving binary codes that exploits their natural matrix structure to reduce their dimensionality using compact bilinear projections instead of a single large projection matrix. This method achieves comparable retrieval and classification accuracy to the original descriptors and to the state-of-the-art Product Quantization approach while having orders of magnitude faster code generation time and smaller memory footprint.

This paper addresses the problem of learning similarity preserving binary codes for efficient retrieval in large-scale image collections. We propose a simple and efficient alternating minimization scheme for finding a rotation of zerocentered data so as to minimize the quantization error of mapping this data to the vertices of a zero-centered binary hypercube. This method, dubbed iterative quantization (ITQ), has connections to multi-class spectral clustering and to the orthogonal Procrustes problem, and it can be used both with unsupervised data embeddings such as PCA and supervised embeddings such as canonical correlation analysis (CCA). Our experiments show that the resulting binary coding schemes decisively outperform several other state-of-the-art methods.

Many approaches to semantic image hashing have been formulated as supervised learning problems that utilize images and label information to learn the binary hash codes. However, large-scale labeled image data is expensive to obtain, thus imposing a restriction on the usage of such algorithms. On the other hand, unlabelled image data is abundant due to the existence of many Web image repositories. Such Web images may often come with images tags that contain useful information, although raw tags, in general, do not readily lead to semantic labels. Motivated by this scenario, we formulate the problem of semantic image hashing as a weakly-supervised learning problem. We utilize the information contained in the user-generated tags associated with the images to learn the hash codes. More specifically, we extract the word2vec semantic embeddings of the tags and use the information contained in them for constraining the learning. Accordingly, we name our model Weakly Supervised Deep Hashing using Tag Embeddings (WDHT). WDHT is tested for the task of semantic image retrieval and is compared against several state-of-art models. Results show that our approach sets a new state-of-art in the area of weekly supervised image hashing.

Learning short binary codes is challenged by the inherent discrete nature of the problem. The graph cuts algorithm is a well-studied discrete label assignment solution in computer vision, but has not yet been applied to solve the binary coding problems. This is partially because it was unclear how to use it to learn the encoding (hashing) functions for out-of-sample generalization. In this paper, we formulate supervised binary coding as a single optimization problem that involves both the encoding functions and the binary label assignment. Then we apply the graph cuts algorithm to address the discrete optimization problem involved, with no continuous relaxation. This method, named as Graph Cuts Coding (GCC), shows competitive results in various datasets.

Approximate nearest neighbor (ANN) query in high-dimensional Euclidean space is a key operator in database systems. For this query, quantization is a popular family of methods developed for compressing vectors and reducing memory consumption. Recently, a method called RaBitQ achieves the state-of-the-art performance among these methods. It produces better empirical performance in both accuracy and efficiency when using the same compression rate and provides rigorous theoretical guarantees. However, the method is only designed for compressing vectors at high compression rates (32x) and lacks support for achieving higher accuracy by using more space. In this paper, we introduce a new quantization method to address this limitation by extending RaBitQ. The new method inherits the theoretical guarantees of RaBitQ and achieves the asymptotic optimality in terms of the trade-off between space and error bounds as to be proven in this study. Additionally, we present efficient implementations of the method, enabling its application to ANN queries to reduce both space and time consumption. Extensive experiments on real-world datasets confirm that our method consistently outperforms the state-of-the-art baselines in both accuracy and efficiency when using the same amount of memory.

Similarity search is a central problem in domains such as information management and retrieval or data analysis. Many similarity search algorithms are designed or specifically adapted to metric distances. Thus, they are unsuitable for alternatives like the cosine distance, which has become quite common, for example, with embeddings and in text mining. This paper presents GDASC (General Distributed Approximate Similarity search with Clustering), a general framework for distributed approximate similarity search that accepts arbitrary distances. This framework can build a multilevel index structure, by selecting a clustering algorithm, the number of prototypes in each cluster and any arbitrary distance function. As a result, this framework effectively overcomes the limitation of using metric distances and can address situations involving cosine similarity or other non-standard similarity measures. Experimental results using k-medoids clustering in GDASC with real datasets confirm the applicability of this approach for approximate similarity search, improving the performance of extant algorithms for this purpose.

In personalized recommender systems, embeddings are often used to encode customer actions and items, and retrieval is then performed in the embedding space using approximate nearest neighbor search. However, this approach can lead to two challenges: 1) user embeddings can restrict the diversity of interests captured and 2) the need to keep them up-to-date requires an expensive, real-time infrastructure. In this paper, we propose a method that overcomes these challenges in a practical, industrial setting. The method dynamically updates customer profiles and composes a feed every two minutes, employing precomputed embeddings and their respective similarities. We tested and deployed this method to personalise promotional items at Bol, one of the largest e-commerce platforms of the Netherlands and Belgium. The method enhanced customer engagement and experience, leading to a significant 4.9% uplift in conversions.

The nearest- or near-neighbor query problems arise in a large variety of database applications, usually in the context of similarity searching. Of late, there has been increasing interest in building search/index structures for performing similarity search over high-dimensional data, e.g., image databases, document collections, time-series databases, and genome databases. Unfortunately, all known techniques for solving this problem fall prey to the curse of dimensionality. That is, the data structures scale poorly with data dimensionality; in fact, if the number of dimensions exceeds 10 to 20, searching in k-d trees and related structures involves the inspection of a large fraction of the database, thereby doing no better than brute-force linear search. It has been suggested that since the selection of features and the choice of a distance metric in typical applications is rather heuristic, determining an approximate nearest neighbor should suffice for most practical purposes. In this paper, we examine a novel scheme for approximate similarity search based on hashing. The basic idea is to hash the points from the database so as to ensure that the probability of collision is much higher for objects that are close to each other than for those that are far apart. We provide experimental evidence that our method gives significant improvement in running time over other methods for searching in highdimensional spaces based on hierarchical tree decomposition. Experimental results also indicate that our scheme scales well even for a relatively large number of dimensions (more than 50).

This paper proposes to learn binary hash codes within a statistical learning framework, in which an upper bound of the probability of Bayes decision errors is derived for different forms of hash functions and a rigorous proof of the convergence of the upper bound is presented. Consequently, minimizing such an upper bound leads to consistent performance improvements of existing hash code learning algorithms, regardless of whether original algorithms are unsupervised or supervised. This paper also illustrates a fast hash coding method that exploits simple binary tests to achieve orders of magnitude improvement in coding speed as compared to projection based methods.

Facebook SimSearchNet++ is a new dataset released by Facebook for this competition. It consists of features used for image copy detection for integrity purposes. The features are generated by Facebook SimSearchNet++ model.

The distribution of keys to a given number of buckets is a fundamental task in distributed data processing and storage. A simple, fast, and therefore popular approach is to map the hash values of keys to buckets based on the remainder after dividing by the number of buckets. Unfortunately, these mappings are not stable when the number of buckets changes, which can lead to severe spikes in system resource utilization, such as network or database requests. Consistent hash algorithms can minimize remappings, but are either significantly slower than the modulo-based approach, require floating-point arithmetic, or are based on a family of hash functions rarely available in standard libraries. This paper introduces JumpBackHash, which uses only integer arithmetic and a standard pseudorandom generator. Due to its speed and simple implementation, it can safely replace the modulo-based approach to improve assignment and system stability. A production-ready Java implementation of JumpBackHash has been released as part of the Hash4j open source library.

This paper proposes a novel deep polarized network (DPN) for learning to hash, in which each channel in the network outputs is pushed far away from zero by employing a differentiable bit-wise hinge-like loss which is dubbed as polarization loss. Reformulated within a generic Hamming Distance Metric Learning framework [Norouzi et al., 2012], the proposed polarization loss bypasses the requirement to prepare pairwise labels for (dis-)similar items and, yet, the proposed loss strictly bounds from above the pairwise Hamming Distance based losses. The intrinsic connection between pairwise and pointwise label information, as disclosed in this paper, brings about the following methodological improvements: (a) we may directly employ the proposed differentiable polarization loss with no large deviations incurred from the target Hamming distance based loss; and (b) the subtask of assigning binary codes becomes extremely simple — even random codes assigned to each class suffice to result in state-of-the-art performances, as demonstrated in CIFAR10, NUS-WIDE and ImageNet100 datasets.

We define and investigate the problem of \(\textit{c-approximate window search}\): approximate nearest neighbor search where each point in the dataset has a numeric label, and the goal is to find nearest neighbors to queries within arbitrary label ranges. Many semantic search problems, such as image and document search with timestamp filters, or product search with cost filters, are natural examples of this problem. We propose and theoretically analyze a modular tree-based framework for transforming an index that solves the traditional c-approximate nearest neighbor problem into a data structure that solves window search. On standard nearest neighbor benchmark datasets equipped with random label values, adversarially constructed embeddings, and image search embeddings with real timestamps, we obtain up to a \(75\times\) speedup over existing solutions at the same level of recall.

The challenge of estimating similarity between sets has been a significant concern in data science, finding diverse applications across various domains. However, previous approaches, such as MinHash, have predominantly centered around hashing techniques, which are well-suited for sets but less naturally adaptable to multisets, a common occurrence in scenarios like network streams and text data. Moreover, with the increasing prevalence of data arriving in streaming patterns, many existing methods struggle to handle cases where set items are presented in a continuous stream. Consequently, our focus in this paper is on the challenging scenario of multisets with item streams. To address this, we propose SimiSketch, a sketching algorithm designed to tackle this specific problem. The paper begins by presenting two simpler versions that employ intuitive sketches for similarity estimation. Subsequently, we formally introduce SimiSketch and leverage SALSA to enhance accuracy. To validate our algorithms, we conduct extensive testing on synthetic datasets, real-world network traffic, and text articles. Our experiment shows that compared with the state-of-the-art, SimiSketch can improve the accuracy by up to 42 times, and increase the throughput by up to 360 times. The complete source code is open-sourced and available on GitHub for reference.

Vector databases typically manage large collections of embedding vectors. Currently, AI applications are growing rapidly, and so is the number of embeddings that need to be stored and indexed. The Faiss library is dedicated to vector similarity search, a core functionality of vector databases. Faiss is a toolkit of indexing methods and related primitives used to search, cluster, compress and transform vectors. This paper describes the trade-off space of vector search and the design principles of Faiss in terms of structure, approach to optimization and interfacing. We benchmark key features of the library and discuss a few selected applications to highlight its broad applicability.

Unsupervised image retrieval aims to learn the important visual characteristics without any given level to retrieve the similar images for a given query image. The Convolutional Neural Network (CNN)-based approaches have been extensively exploited with self-supervised contrastive learning for image hashing. However, the existing approaches suffer due to lack of effective utilization of global features by CNNs and biased-ness created by false negative pairs in the contrastive learning. In this paper, we propose a TransClippedCLR model by encoding the global context of an image using Transformer having local context through patch based processing, by generating the hash codes through product quantization and by avoiding the potential false negative pairs through clipped contrastive learning. The proposed model is tested with superior performance for unsupervised image retrieval on benchmark datasets, including CIFAR10, NUS-Wide and Flickr25K, as compared to the recent state-of-the-art deep models. The results using the proposed clipped contrastive learning are greatly improved on all datasets as compared to same backbone network with vanilla contrastive learning.

Hashing is very effective for many tasks in reducing the processing time and in compressing massive databases. Although lots of approaches have been developed to learn data-dependent hash functions in recent years, how to learn hash functions to yield good performance with acceptable computational and memory cost is still a challenging problem. Based on the observation that retrieval precision is highly related to the kNN classification accuracy, this paper proposes a novel kNN-based supervised hashing method, which learns hash functions by directly maximizing the kNN accuracy of the Hamming-embedded training data. To make it scalable well to large problem, we propose a factorized neighborhood representation to parsimoniously model the neighborhood relationships inherent in training data. Considering that real-world data are often linearly inseparable, we further kernelize this basic model to improve its performance. As a result, the proposed method is able to learn accurate hashing functions with tolerable computation and storage cost. Experiments on four benchmarks demonstrate that our method outperforms the state-of-the-arts.

Retrieval plays a fundamental role in recommendation systems, search, and natural language processing by efficiently finding relevant items from a large corpus given a query. Dot products have been widely used as the similarity function in such retrieval tasks, thanks to Maximum Inner Product Search (MIPS) that enabled efficient retrieval based on dot products. However, state-of-the-art retrieval algorithms have migrated to learned similarities. Such algorithms vary in form; the queries can be represented with multiple embeddings, complex neural networks can be deployed, the item ids can be decoded directly from queries using beam search, and multiple approaches can be combined in hybrid solutions. Unfortunately, we lack efficient solutions for retrieval in these state-of-the-art setups. Our work investigates techniques for approximate nearest neighbor search with learned similarity functions. We first prove that Mixture-of-Logits (MoL) is a universal approximator, and can express all learned similarity functions. We next propose techniques to retrieve the approximate top K results using MoL with a tight bound. We finally compare our techniques with existing approaches, showing that MoL sets new state-of-the-art results on recommendation retrieval tasks, and our approximate top-k retrieval with learned similarities outperforms baselines by up to two orders of magnitude in latency, while achieving > .99 recall rate of exact algorithms.

The \(k\)-Minimum Values (\kmv) data sketch algorithm stores the \(k\) least hash keys generated by hashing the items in a dataset. We show that compression based on ordering the keys and encoding successive differences can offer \(O(log n)\) bits per key in expected storage savings, where \(n\) is the number of unique values in the data set. We also show that \(O(log n)\) expected bits saved per key is optimal for any form of compression for the \(k\) least of \(n\) random values – that the encoding method is near-optimal among all methods to encode a \kmv sketch. We present a practical method to perform that compression, show that it is computationally efficient, and demonstrate that its average savings in practice is within about five percent of the theoretical minimum based on entropy. We verify that our method outperforms off-the-shelf compression methods, and we demonstrate that it is practical, using real and synthetic data.

Gene sequence search is a fundamental operation in computational genomics. Due to the petabyte scale of genome archives, most gene search systems now use hashing-based data structures such as Bloom Filters (BF). The state-of-the-art systems such as Compact bit-slicing signature index (COBS) and Repeated And Merged Bloom filters (RAMBO) use BF with Random Hash (RH) functions for gene representation and identification. The standard recipe is to cast the gene search problem as a sequence of membership problems testing if each subsequent gene substring (called kmer) of Q is present in the set of kmers of the entire gene database D. We observe that RH functions, which are crucial to the memory and the computational advantage of BF, are also detrimental to the system performance of gene-search systems. While subsequent kmers being queried are likely very similar, RH, oblivious to any similarity, uniformly distributes the kmers to different parts of potentially large BF, thus triggering excessive cache misses and causing system slowdown. We propose a novel hash function called the Identity with Locality (IDL) hash family, which co-locates the keys close in input space without causing collisions. This approach ensures both cache locality and key preservation. IDL functions can be a drop-in replacement for RH functions and help improve the performance of information retrieval systems. We give a simple but practical construction of IDL function families and show that replacing the RH with IDL functions reduces cache misses by a factor of 5x, thus improving query and indexing times of SOTA methods such as COBS and RAMBO by factors up to 2x without compromising their quality. We also provide a theoretical analysis of the false positive rate of BF with IDL functions. Our hash function is the first study that bridges Locality Sensitive Hash (LSH) and RH to obtain cache efficiency.

Neural embedding models have become a fundamental component of modern information retrieval (IR) pipelines. These models produce a single embedding \(x \in \mathbb{R}^d\) per data-point, allowing for fast retrieval via highly optimized maximum inner product search (MIPS) algorithms. Recently, beginning with the landmark ColBERT paper, multi-vector models, which produce a set of embedding per data point, have achieved markedly superior performance for IR tasks. Unfortunately, using these models for IR is computationally expensive due to the increased complexity of multi-vector retrieval and scoring. In this paper, we introduce MUVERA (MUlti-VEctor Retrieval Algorithm), a retrieval mechanism which reduces multi-vector similarity search to single-vector similarity search. This enables the usage of off-the-shelf MIPS solvers for multi-vector retrieval. MUVERA asymmetrically generates Fixed Dimensional Encodings (FDEs) of queries and documents, which are vectors whose inner product approximates multi-vector similarity. We prove that FDEs give high-quality \(\epsilon\)-approximations, thus providing the first single-vector proxy for multi-vector similarity with theoretical guarantees. Empirically, we find that FDEs achieve the same recall as prior state-of-the-art heuristics while retrieving 2-5\(\times\) fewer candidates. Compared to prior state of the art implementations, MUVERA achieves consistently good end-to-end recall and latency across a diverse set of the BEIR retrieval datasets, achieving an average of 10\(\%\) improved recall with \(90\%\) lower latency.

Nearest neighbor search methods based on hashing have attracted considerable attention for effective and efficient large-scale similarity search in computer vision and information retrieval community. In this paper, we study the problems of learning hash functions in the context of multimodal data for cross-view similarity search. We put forward a novel hashing method, which is referred to Collective Matrix Factorization Hashing (CMFH). CMFH learns unified hash codes by collective matrix factorization with latent factor model from different modalities of one instance, which can not only supports cross-view search but also increases the search accuracy by merging multiple view information sources. We also prove that CMFH, a similaritypreserving hashing learning method, has upper and lower boundaries. Extensive experiments verify that CMFH significantly outperforms several state-of-the-art methods on three different datasets.

Space partitions of underlie a vast and important class of fast nearest neighbor search (NNS) algorithms. Inspired by recent theoretical work on NNS for general metric spaces (Andoni et al. 2018b,c), we develop a new framework for building space partitions reducing the problem to balanced graph partitioning followed by supervised classification. We instantiate this general approach with the KaHIP graph partitioner (Sanders and Schulz 2013) and neural networks, respectively, to obtain a new partitioning procedure called Neural Locality-Sensitive Hashing (Neural LSH). On several standard benchmarks for NNS (Aumuller et al. 2017), our experiments show that the partitions obtained by Neural LSH consistently outperform partitions found by quantization-based and tree-based methods as well as classic, data-oblivious LSH.

We present a novel Locality-Sensitive Hashing scheme for the Approximate Nearest Neighbor Problem under lp norm, based on p-stable distributions.Our scheme improves the running time of the earlier algorithm for the case of the lp norm. It also yields the first known provably efficient approximate NN algorithm for the case p<1. We also show that the algorithm finds the exact near neigbhor in O(log n) time for data satisfying certain “bounded growth” condition.Unlike earlier schemes, our LSH scheme works directly on points in the Euclidean space without embeddings. Consequently, the resulting query time bound is free of large factors and is simple and easy to implement. Our experiments (on synthetic data sets) show that the our data structure is up to 40 times faster than kd-tree.

The explosion of image data on the Internet has the potential to foster more sophisticated and robust models and algorithms to index, retrieve, organize and interact with images and multimedia data. But exactly how such data can be harnessed and organized remains a critical problem. We introduce here a new database called “ImageNet”, a large-scale ontology of images built upon the backbone of the WordNet structure. ImageNet aims to populate the majority of the 80,000 synsets of WordNet with an average of 500-1000 clean and full resolution images. This will result in tens of millions of annotated images organized by the semantic hierarchy of WordNet. This paper offers a detailed analysis of ImageNet in its current state: 12 subtrees with 5247 synsets and 3.2 million images in total. We show that ImageNet is much larger in scale and diversity and much more accurate than the current image datasets. Constructing such a large-scale database is a challenging task. We describe the data collection scheme with Amazon Mechanical Turk. Lastly, we illustrate the usefulness of ImageNet through three simple applications in object recognition, image classification and automatic object clustering. We hope that the scale, accuracy, diversity and hierarchical structure of ImageNet can offer unparalleled opportunities to researchers in the computer vision community and beyond.

Retrieving content relevant images from a large-scale fine grained dataset could suffer from intolerably slow query speed and highly redundant storage cost, due to high-dimensional real-valued embeddings which aim to distinguish subtle visual differences of fine-grained objects. In this paper, we study the novel fine-grained hashing topic to generate compact binary codes for fine-grained images, leveraging the search and storage efficiency of hash learning to alleviate the aforementioned problems. Specifically, we propose a unified end-to-end trainable network, termed as ExchNet. Based on attention mechanisms and proposed attention constraints, it can firstly obtain both local and global features to represent object parts and whole fine-grained objects, respectively. Furthermore, to ensure the discriminative ability and semantic meaning’s consistency of these part-level features across images, we design a local feature alignment approach by performing a feature exchanging operation. Later, an alternative learning algorithm is employed to optimize the whole ExchNet and then generate the final binary hash codes. Validated by extensive experiments, our proposal consistently outperforms state-of-the-art generic hashing methods on five fine-grained datasets, which shows our effectiveness. Moreover, compared with other approximate nearest neighbor methods, ExchNet achieves the best speed-up and storage reduction, revealing its efficiency and practicality.

Consistent hashing is employed in distributed systems and networking applications to evenly and effectively distribute data across a cluster of nodes. This paper introduces BinomialHash, a consistent hashing algorithm that operates in constant time and requires minimal memory. We provide a detailed explanation of the algorithm, offer a pseudo-code implementation, and formally establish its strong theoretical guarantees.

Binary code similarity detection is an important problem with applications in areas like malware analysis, vulnerability research and plagiarism detection. This paper proposes a novel graph neural network architecture combined with a novel graph data representation called call graphlets. A call graphlet encodes the neighborhood around each function in a binary executable, capturing the local and global context through a series of statistical features. A specialized graph neural network model is then designed to operate on this graph representation, learning to map it to a feature vector that encodes semantic code similarities using deep metric learning. The proposed approach is evaluated across four distinct datasets covering different architectures, compiler toolchains, and optimization levels. Experimental results demonstrate that the combination of call graphlets and the novel graph neural network architecture achieves state-of-the-art performance compared to baseline techniques across cross-architecture, mono-architecture and zero shot tasks. In addition, our proposed approach also performs well when evaluated against an out-of-domain function inlining task. Overall, the work provides a general and effective graph neural network-based solution for conducting binary code similarity detection.

Hashing has been widely used for large-scale approximate nearest neighbor search due to its storage and search efficiency. Recent supervised hashing research has shown that deep learning-based methods can significantly outperform nondeep methods. Most existing supervised deep hashing methods exploit supervisory signals to generate similar and dissimilar image pairs for training. However, natural images can have large intraclass and small interclass variations, which may degrade the accuracy of hash codes. To address this problem, we propose a novel two-stream ConvNet architecture, which learns hash codes with class-specific representation centers. Our basic idea is that if we can learn a unified binary representation for each class as a center and encourage hash codes of images to be close to the corresponding centers, the intraclass variation will be greatly reduced. Accordingly, we design a neural network that leverages label information and outputs a unified binary representation for each class. Moreover, we also design an image network to learn hash codes from images and force these hash codes to be close to the corresponding class-specific centers. These two neural networks are then seamlessly incorporated to create a unified, end-to-end trainable framework. Extensive experiments on three popular benchmarks corroborate that our proposed method outperforms current state-of-the-art methods.

This paper introduces a web image dataset created by NUS’s Lab for Media Search. The dataset includes: (1) 269,648 images and the associated tags from Flickr, with a total of 5,018 unique tags; (2) six types of low-level features extracted from these images, including 64-D color histogram, 144-D color correlogram, 73-D edge direction histogram, 128-D wavelet texture, 225-D block-wise color moments extracted over 5x5 fixed grid partitions, and 500-D bag of words based on SIFT descriptions; and (3) ground-truth for 81 concepts that can be used for evaluation. Based on this dataset, we highlight characteristics of Web image collections and identify four research issues on web image annotation and retrieval. We also provide the baseline results for web image annotation by learning from the tags using the traditional k-NN algorithm. The benchmark results indicate that it is possible to learn effective models from sufficiently large image dataset to facilitate general image retrieval.

In this paper, we present a novel Two-stEp Cross-modal Hashing method, TECH for short, for cross-modal retrieval tasks. As a two-step method, it first learns hash codes based on semantic labels, while preserving the similarity in the original space and exploiting the label correlations in the label space. In the light of this, it is able to make better use of label information and generate better binary codes. In addition, different from other two-step methods that mainly focus on the hash codes learning, TECH adopts a new hash function learning strategy in the second step, which also preserves the similarity in the original space. Moreover, with the help of well designed objective function and optimization scheme, it is able to generate hash codes discretely and scalable for large scale data. To the best of our knowledge, it is the first cross-modal hashing method exploiting label correlations, and also the first two-step hashing model preserving the similarity while leaning hash function. Extensive experiments demonstrate that the proposed approach outperforms some state-of-the-art cross-modal hashing methods.

With the quick development of social websites, there are more opportunities to have different media types (such as text, image, video, etc.) describing the same topic from large-scale heterogeneous data sources. To efficiently identify the inter-media correlations for multimedia retrieval, unsupervised cross-modal hashing (UCMH) has gained increased interest due to the significant reduction in computation and storage. However, most UCMH methods assume that the data from different modalities are well paired. As a result, existing UCMH methods may not achieve satisfactory performance when partially paired data are given only. In this article, we propose a new-type of UCMH method called robust unsupervised cross-modal hashing (RUCMH). The major contribution lies in jointly learning modal-specific hash function, exploring the correlations among modalities with partial or even without any pairwise correspondence, and preserving the information of original features as much as possible. The learning process can be modeled via a joint minimization problem, and the corresponding optimization algorithm is presented. A series of experiments is conducted on four real-world datasets (Wiki, MIRFlickr, NUS-WIDE, and MS-COCO). The results demonstrate that RUCMH can significantly outperform the state-of-the-art unsupervised cross-modal hashing methods, especially for the partially paired case, which validates the effectiveness of RUCMH.

The target of cross-modal hashing is to embed heterogeneous multimedia data into a common low-dimensional Hamming space, which plays a pivotal part in multimedia retrieval due to the emergence of big multimodal data. Recently, matrix factorization has achieved great success in cross-modal hashing. However, how to effectively use label information and local geometric structure is still a challenging problem for these approaches. To address this issue, we propose a cross-modal hashing method based on collective matrix factorization, which considers both the label consistency across different modalities and the local geometric consistency in each modality. These two elements are formulated as a graph Laplacian term in the objective function, leading to a substantial improvement on the discriminative power of latent semantic features obtained by collective matrix factorization. Moreover, the proposed method learns unified hash codes for different modalities of an instance to facilitate cross-modal search, and the objective function is solved using an iterative strategy. The experimental results on two benchmark data sets show the effectiveness of the proposed method and its superiority over state-of-the-art cross-modal hashing methods.

Hashing, which represents data items as compact binary codes, has been becoming a more and more popular technique, e.g., for large-scale image retrieval, owing to its super fast search speed as well as its extremely economical memory consumption. However, existing hashing methods all try to learn binary codes from artificially balanced datasets which are not commonly available in real-world scenarios. In this paper, we propose Long-Tail Hashing Network (LTHNet), a novel two-stage deep hashing approach that addresses the problem of learning to hash for more realistic datasets where the data labels roughly exhibit a long-tail distribution. Specifically, the first stage is to learn relaxed embeddings of the given dataset with its long-tail characteristic taken into account via an end-to-end deep neural network; the second stage is to binarize those obtained embeddings. A critical part of LTHNet is its extended dynamic meta-embedding module which can adaptively realize visual knowledge transfer between head and tail classes, and thus enrich image representations for hashing. Our experiments have shown that LTHNet achieves dramatic performance improvements over all state-of-the-art competitors on long-tail datasets, with no or little sacrifice on balanced datasets. Further analyses reveal that while to our surprise directly manipulating class weights in the loss function has little effect, the extended dynamic meta-embedding module, the usage of cross-entropy loss instead of square loss, and the relatively small batch-size for training all contribute to LTHNet’s success.

Computing approximate nearest neighbors in high dimensional spaces is a central problem in large-scale data mining with a wide range of applications in machine learning and data science. A popular and effective technique in computing nearest neighbors approximately is the locality-sensitive hashing (LSH) scheme. In this paper, we aim to develop LSH schemes for distance functions that measure the distance between two probability distributions, particularly for f-divergences as well as a generalization to capture mutual information loss. First, we provide a general framework to design LHS schemes for f-divergence distance functions and develop LSH schemes for the generalized Jensen-Shannon divergence and triangular discrimination in this framework. We show a two-sided approximation result for approximation of the generalized Jensen-Shannon divergence by the Hellinger distance, which may be of independent interest. Next, we show a general method of reducing the problem of designing an LSH scheme for a Krein kernel (which can be expressed as the difference of two positive definite kernels) to the problem of maximum inner product search. We exemplify this method by applying it to the mutual information loss, due to its several important applications such as model compression.

Learning to hash is a fundamental technique widely used in large-scale image retrieval. Most existing methods for learning to hash address the involved discrete optimization problem by the continuous relaxation of the binary constraint, which usually leads to large quantization errors and consequently suboptimal binary codes. A few discrete hashing methods have emerged recently. However, they either completely ignore some useful constraints (specifically the balance and decorrelation of hash bits) or just turn those constraints into regularizers that would make the optimization easier but less accurate. In this paper, we propose a novel supervised hashing method named Strongly Constrained Discrete Hashing (SCDH) which overcomes such limitations. It can learn the binary codes for all examples in the training set, and meanwhile obtain a hash function for unseen samples with the above mentioned constraints preserved. Although the model of SCDH is fairly sophisticated, we are able to find closed-form solutions to all of its optimization subproblems and thus design an efficient algorithm that converges quickly. In addition, we extend SCDH to a kernelized version SCDH K . Our experiments on three large benchmark datasets have demonstrated that not only can SCDH and SCDH K achieve substantially higher MAP scores than state-of-the-art baselines, but they train much faster than those that are also supervised as well.

Searching on bipartite graphs serves as a fundamental task for various real-world applications, such as recommendation systems, database retrieval, and document querying. Conventional approaches rely on similarity matching in continuous Euclidean space of vectorized node embeddings. To handle intensive similarity computation efficiently, hashing techniques for graph-structured data have emerged as a prominent research direction. However, despite the retrieval efficiency in Hamming space, previous studies have encountered catastrophic performance decay. To address this challenge, we investigate the problem of hashing with Graph Convolutional Network for effective Top-N search. Our findings indicate the learning effectiveness of incorporating hashing techniques within the exploration of bipartite graph reception fields, as opposed to simply treating hashing as post-processing to output embeddings. To further enhance the model performance, we advance upon these findings and propose Bipartite Graph Contrastive Hashing (BGCH+). BGCH+ introduces a novel dual augmentation approach to both intermediate information and hash code outputs in the latent feature spaces, thereby producing more expressive and robust hash codes within a dual self-supervised learning paradigm. Comprehensive empirical analyses on six real-world benchmarks validate the effectiveness of our dual feature contrastive learning in boosting the performance of BGCH+ compared to existing approaches.

Recently, a series of deep supervised hashing methods were proposed for binary code learning. However, due to the high computation cost and the limited hardware’s memory, these methods will first select a subset from the training set, and then form a mini-batch data to update the network in each iteration. Therefore, the remaining labeled data cannot be fully utilized and the model cannot directly obtain the binary codes of the entire training set for retrieval. To address these problems, this paper proposes an interesting regularized deep model to seamlessly integrate the advantages of deep hashing and efficient binary code learning by using the anchor graph. As such, the deep features and label matrix can be jointly used to optimize the binary codes, and the network can obtain more discriminative feedback from the linear combinations of the learned bits. Moreover, we also reveal the algorithm mechanism and its computation essence. Experiments on three large-scale datasets indicate that the proposed method achieves better retrieval performance with less training time compared to previous deep hashing methods.

Due to the exponential growth of multimedia data, multi-modal hashing as a promising technique to make cross-view retrieval scalable is attracting more and more attention. However, most of the existing multi-modal hashing methods either divide the learning process unnaturally into two separate stages or treat the discrete optimization problem simplistically as a continuous one, which leads to suboptimal results. Recently, a few discrete multi-modal hashing methods that try to address such issues have emerged, but they still ignore several important discrete constraints (such as the balance and decorrelation of hash bits). In this paper, we overcome those limitations by proposing a novel method named “Enhanced Discrete Multi-modal Hashing (EDMH)” which learns binary codes and hashing functions simultaneously from the pairwise similarity matrix of data, under the aforementioned discrete constraints. Although the model of EDMH looks a lot more complex than the other models for multi-modal hashing, we are actually able to develop a fast iterative learning algorithm for it, since the subproblems of its optimization all have closed-form solutions after introducing two auxiliary variables. Our experimental results on three real-world datasets have demonstrated that EDMH not only performs much better than state-of-the-art competitors but also runs much faster than them.

3D Gaussian Splatting (3DGS) has emerged as a promising framework for novel view synthesis, boasting rapid rendering speed with high fidelity. However, the substantial Gaussians and their associated attributes necessitate effective compression techniques. Nevertheless, the sparse and unorganized nature of the point cloud of Gaussians (or anchors in our paper) presents challenges for compression. To address this, we make use of the relations between the unorganized anchors and the structured hash grid, leveraging their mutual information for context modeling, and propose a Hash-grid Assisted Context (HAC) framework for highly compact 3DGS representation. Our approach introduces a binary hash grid to establish continuous spatial consistencies, allowing us to unveil the inherent spatial relations of anchors through a carefully designed context model. To facilitate entropy coding, we utilize Gaussian distributions to accurately estimate the probability of each quantized attribute, where an adaptive quantization module is proposed to enable high-precision quantization of these attributes for improved fidelity restoration. Additionally, we incorporate an adaptive masking strategy to eliminate invalid Gaussians and anchors. Importantly, our work is the pioneer to explore context-based compression for 3DGS representation, resulting in a remarkable size reduction of over \(75\times\) compared to vanilla 3DGS, while simultaneously improving fidelity, and achieving over \(11\times\) size reduction over SOTA 3DGS compression approach Scaffold-GS. Our code is available here: https://github.com/YihangChen-ee/HAC

With an ever increasing amount of data available on the web, fast similarity search has become the critical component for large-scale information retrieval systems. One solution is semantic hashing which designs binary codes to accelerate similarity search. Recently, deep learning has been successfully applied to the semantic hashing problem and produces high-quality compact binary codes compared to traditional methods. However, most state-of-the-art semantic hashing approaches require large amounts of hand-labeled training data which are often expensive and time consuming to collect. The cost of getting labeled data is the key bottleneck in deploying these hashing methods. Motivated by the recent success in machine learning that makes use of weak supervision, we employ unsupervised ranking methods such as BM25 to extract weak signals from training data. We further introduce two deep generative semantic hashing models to leverage weak signals for text hashing. The experimental results on four public datasets show that our models can generate high-quality binary codes without using hand-labeled training data and significantly outperform the competitive unsupervised semantic hashing baselines.

As the amount of textual data has been rapidly increasing over the past decade, efficient similarity search methods have become a crucial component of large-scale information retrieval systems. A popular strategy is to represent original data samples by compact binary codes through hashing. A spectrum of machine learning methods have been utilized, but they often lack expressiveness and flexibility in modeling to learn effective representations. The recent advances of deep learning in a wide range of applications has demonstrated its capability to learn robust and powerful feature representations for complex data. Especially, deep generative models naturally combine the expressiveness of probabilistic generative models with the high capacity of deep neural networks, which is very suitable for text modeling. However, little work has leveraged the recent progress in deep learning for text hashing. In this paper, we propose a series of novel deep document generative models for text hashing. The first proposed model is unsupervised while the second one is supervised by utilizing document labels/tags for hashing. The third model further considers document-specific factors that affect the generation of words. The probabilistic generative formulation of the proposed models provides a principled framework for model extension, uncertainty estimation, simulation, and interpretability. Based on variational inference and reparameterization, the proposed models can be interpreted as encoder-decoder deep neural networks and thus they are capable of learning complex nonlinear distributed representations of the original documents. We conduct a comprehensive set of experiments on four public testbeds. The experimental results have demonstrated the effectiveness of the proposed supervised learning models for text hashing.

An attractive approach for fast search in image databases is binary hashing, where each high-dimensional, real-valued image is mapped onto a low-dimensional, binary vector and the search is done in this binary space. Finding the optimal hash function is difficult because it involves binary constraints, and most approaches approximate the optimization by relaxing the constraints and then binarizing the result. Here, we focus on the binary autoencoder model, which seeks to reconstruct an image from the binary code produced by the hash function. We show that the optimization can be simplified with the method of auxiliary coordinates. This reformulates the optimization as alternating two easier steps: one that learns the encoder and decoder separately, and one that optimizes the code for each image. Image retrieval experiments show the resulting hash function outperforms or is competitive with state-ofthe-art methods for binary hashing.

Due to its computation efficiency and retrieval quality, hashing has been widely applied to approximate nearest neighbor search for large-scale image retrieval, while deep hashing further improves the retrieval quality by end-toend representation learning and hash coding. With compact hash codes, Hamming space retrieval enables the most efficient constant-time search that returns data points within a given Hamming radius to each query, by hash table lookups instead of linear scan. However, subject to the weak capability of concentrating relevant images to be within a small Hamming ball due to mis-specified loss functions, existing deep hashing methods may underperform for Hamming space retrieval. This work presents Deep Cauchy Hashing (DCH), a novel deep hashing model that generates compact and concentrated binary hash codes to enable efficient and effective Hamming space retrieval. The main idea is to design a pairwise cross-entropy loss based on Cauchy distribution, which penalizes significantly on similar image pairs with Hamming distance larger than the given Hamming radius threshold. Comprehensive experiments demonstrate that DCH can generate highly concentrated hash codes and yield state-of-the-art Hamming space retrieval performance on three datasets, NUS-WIDE, CIFAR-10, and MS-COCO.

Due to its storage and query efficiency, hashing has been widely applied to approximate nearest neighbor search from large-scale datasets. While there is increasing interest in cross-modal hashing which facilitates cross-media retrieval by embedding data from different modalities into a common Hamming space, how to distill the cross-modal correlation structure effectively remains a challenging problem. In this paper, we propose a novel supervised cross-modal hashing method, Correlation Autoencoder Hashing (CAH), to learn discriminative and compact binary codes based on deep autoencoders. Specifically, CAH jointly maximizes the feature correlation revealed by bimodal data and the semantic correlation conveyed in similarity labels, while embeds them into hash codes by nonlinear deep autoencoders. Extensive experiments clearly show the superior effectiveness and efficiency of CAH against the state-of-the-art hashing methods on standard cross-modal retrieval benchmarks.

Deep learning to hash improves image retrieval performance by end-to-end representation learning and hash coding from training data with pairwise similarity information. Subject to the scarcity of similarity information that is often expensive to collect for many application domains, existing deep learning to hash methods may overfit the training data and result in substantial loss of retrieval quality. This paper presents HashGAN, a novel architecture for deep learning to hash, which learns compact binary hash codes from both real images and diverse images synthesized by generative models. The main idea is to augment the training data with nearly real images synthesized from a new Pair Conditional Wasserstein GAN (PC-WGAN) conditioned on the pairwise similarity information. Extensive experiments demonstrate that HashGAN can generate high-quality binary hash codes and yield state-of-the-art image retrieval performance on three benchmarks, NUS-WIDE, CIFAR-10, and MS-COCO.

Cross-modal similarity retrieval is a problem about designing a retrieval system that supports querying across content modalities, e.g., using an image to retrieve for texts. This paper presents a compact coding solution for efficient cross-modal retrieval, with a focus on the quantization approach which has already shown the superior performance over the hashing solutions in single-modal similarity retrieval. We propose a collective deep quantization (CDQ) approach, which is the first attempt to introduce quantization in end-to-end deep architecture for cross-modal retrieval. The major contribution lies in jointly learning deep representations and the quantizers for both modalities using carefully-crafted hybrid networks and well-specified loss functions. In addition, our approach simultaneously learns the common quantizer codebook for both modalities through which the crossmodal correlation can be substantially enhanced. CDQ enables efficient and effective cross-modal retrieval using inner product distance computed based on the common codebook with fast distance table lookup. Extensive experiments show that CDQ yields state of the art cross-modal retrieval results on standard benchmarks.

Learning to hash has been widely applied to approximate nearest neighbor search for large-scale multimedia retrieval, due to its computation efficiency and retrieval quality. Deep learning to hash, which improves retrieval quality by end-to-end representation learning and hash encoding, has received increasing attention recently. Subject to the illposed gradient difficulty in the optimization with sign activations, existing deep learning to hash methods need to first learn continuous representations and then generate binary hash codes in a separated binarization step, which suffer from substantial loss of retrieval quality. This work presents HashNet, a novel deep architecture for deep learning to hash by continuation method with convergence guarantees, which learns exactly binary hash codes from imbalanced similarity data. The key idea is to attack the ill-posed gradient problem in optimizing deep networks with non-smooth binary activations by continuation method, in which we begin from learning an easier network with smoothed activation function and let it evolve during the training, until it eventually goes back to being the original, difficult to optimize, deep network with the sign activation function. Comprehensive empirical evidence shows that HashNet can generate exactly binary hash codes and yield state-of-the-art multimedia retrieval performance on standard benchmarks.

With the staggering growth in image and video datasets, algorithms that provide fast similarity search and compact storage are crucial. Hashing methods that map the data into Hamming space have shown promise; however, many of these methods employ a batch-learning strategy in which the computational cost and memory requirements may become intractable and infeasible with larger and larger datasets. To overcome these challenges, we propose an online learning algorithm based on stochastic gradient descent in which the hash functions are updated iteratively with streaming data. In experiments with three image retrieval benchmarks, our online algorithm attains retrieval accuracy that is comparable to competing state-of-the-art batch-learning solutions, while our formulation is orders of magnitude faster and being online it is adaptable to the variations of the data. Moreover, our formulation yields improved retrieval performance over a recently reported online hashing technique, Online Kernel Hashing.

We propose theoretical and empirical improvements for two-stage hashing methods. We first provide a theoretical analysis on the quality of the binary codes and show that, under mild assumptions, a residual learning scheme can construct binary codes that fit any neighborhood structure with arbitrary accuracy. Secondly, we show that with high-capacity hash functions such as CNNs, binary code inference can be greatly simplified for many standard neighborhood definitions, yielding smaller optimization problems and more robust codes. Incorporating our findings, we propose a novel two-stage hashing method that significantly outperforms previous hashing studies on widely used image retrieval benchmarks.

Learned dense representations are a popular family of techniques for encoding queries and documents using high-dimensional embeddings, which enable retrieval by performing approximate k nearest-neighbors search (A-kNN). A popular technique for making A-kNN search efficient is based on a two-level index, where the embeddings of documents are clustered offline and, at query processing, a fixed number N of clusters closest to the query is visited exhaustively to compute the result set. In this paper, we build upon state-of-the-art for early exit A-kNN and propose an unsupervised method based on the notion of patience, which can reach competitive effectiveness with large efficiency gains. Moreover, we discuss a cascade approach where we first identify queries that find their nearest neighbor within the closest t « N clusters, and then we decide how many more to visit based on our patience approach or other state-of-the-art strategies. Reproducible experiments employing state-of-the-art dense retrieval models and publicly available resources show that our techniques improve the A-kNN efficiency with up to 5x speedups while achieving negligible effectiveness losses. All the code used is available at https://github.com/francescobusolin/faiss_pEE

Learning-based hashing methods are widely used for nearest neighbor retrieval, and recently, online hashing methods have demonstrated good performance-complexity trade-offs by learning hash functions from streaming data. In this paper, we first address a key challenge for online hashing: the binary codes for indexed data must be recomputed to keep pace with updates to the hash functions. We propose an efficient quality measure for hash functions, based on an information-theoretic quantity, mutual information, and use it successfully as a criterion to eliminate unnecessary hash table updates. Next, we also show how to optimize the mutual information objective using stochastic gradient descent. We thus develop a novel hashing method, MIHash, that can be used in both online and batch settings. Experiments on image retrieval benchmarks (including a 2.5M image dataset) confirm the effectiveness of our formulation, both in reducing hash table recomputations and in learning high-quality hash functions.

We define and study the notion of min-wise independent families of permutations. Our research was motivated by the fact that such a family (under some relaxations) is essential to the algorithm used in practice by the AltaVista web index software to detect and filter near-duplicate documents. However, in the course of our investigation we have discovered interesting and challenging theoretical questions related to this concept we present the solutions to some of them and we list the rest as open problems.

We consider the problem of indexing high-dimensional data for answering (approximate) similarity-search queries. Similarity indexes prove to be important in a wide variety of settings: Web search engines desire fast, parallel, main-memory-based indexes for similarity search on text data; database systems desire disk-based similarity indexes for high-dimensional data, including text and images; peer-to-peer systems desire distributed similarity indexes with low communication cost. We propose an indexing scheme called LSH Forest which is applicable in all the above contexts. Our index uses the well-known technique of locality-sensitive hashing (LSH), but improves upon previous designs by (a) eliminating the different data-dependent parameters for which LSH must be constantly hand-tuned, and (b) improving on LSH’s performance guarantees for skewed data distributions while retaining the same storage and query overhead. We show how to construct this index in main memory, on disk, in parallel systems, and in peer-to-peer systems. We evaluate the design with experiments on multiple text corpora and demonstrate both the self-tuning nature and the superior performance of LSH Forest.

The deep hashing based retrieval method is widely adopted in large-scale image and video retrieval. However, there is little investigation on its security. In this paper, we propose a novel method, dubbed deep hashing targeted attack (DHTA), to study the targeted attack on such retrieval. Specifically, we first formulate the targeted attack as a point-to-set optimization, which minimizes the average distance between the hash code of an adversarial example and those of a set of objects with the target label. Then we design a novel component-voting scheme to obtain an anchor code as the representative of the set of hash codes of objects with the target label, whose optimality guarantee is also theoretically derived. To balance the performance and perceptibility, we propose to minimize the Hamming distance between the hash code of the adversarial example and the anchor code under the ℓ∞ restriction on the perturbation. Extensive experiments verify that DHTA is effective in attacking both deep hashing based image retrieval and video retrieval.

The expanse of information available over the internet makes it difficult to identify whether a specific work is a replica or a duplication of a protected work, especially if we talk about visual representations. Strategies are planned to identify the utilization of the copyrighted image in a report. Still, we want to resolve the issue of involving a copyrighted image in a video and a calculation that could recognize the degree of similarity of the copyrighted picture utilized in the video, even for the pieces of the video that are not featured a lot and in the end perform characterization errands on those edges. Machine learning (ML) and artificial intelligence (AI) are vital to address this problem. Numerous associations have been creating different calculations to screen the identification of copyrighted work. This work means concentrating on those calculations, recognizing designs inside the information, and fabricating a more reasonable model for copyrighted image classification and detection. We have used different algorithms like- Image Processing, Convolutional Neural Networks (CNN), Image hashing, etc. Keywords- Copyright, Artificial Intelligence(AI), Copyrighted Image, Convolutional Neural Network(CNN), Image processing, Degree of similarity, Image Hashing.

We present an algorithm for the c-approximate nearest neighbor problem in a d-dimensional Euclidean space, achieving query time of O(dn 1c2/+o(1)) and space O(dn + n1+1c2/+o(1)). This almost matches the lower bound for hashing-based algorithm recently obtained in (R. Motwani et al., 2006). We also obtain a space-efficient version of the algorithm, which uses dn+n logO(1) n space, with a query time of dnO(1/c2). Finally, we discuss practical variants of the algorithms that utilize fast bounded-distance decoders for the Leech lattice

The indexing algorithms for the high-dimensional nearest neighbor search (NNS) with the best worst-case guarantees are based on the randomized Locality Sensitive Hashing (LSH), and its derivatives. In practice, many heuristic approaches exist to “learn” the best indexing method in order to speed-up NNS, crucially adapting to the structure of the given dataset. Oftentimes, these heuristics outperform the LSH-based algorithms on real datasets, but, almost always, come at the cost of losing the guarantees of either correctness or robust performance on adversarial queries, or apply to datasets with an assumed extra structure/model. In this paper, we design an NNS algorithm for the Hamming space that has worst-case guarantees essentially matching that of theoretical algorithms, while optimizing the hashing to the structure of the dataset (think instance-optimal algorithms) for performance on the minimum-performing query. We evaluate the algorithm’s ability to optimize for a given dataset both theoretically and practically. On the theoretical side, we exhibit a natural setting (dataset model) where our algorithm is much better than the standard theoretical one. On the practical side, we run experiments that show that our algorithm has a 1.8x and 2.1x better recall on the worst-performing queries to the MNIST and ImageNet datasets.

We show the existence of a Locality-Sensitive Hashing (LSH) family for the angular distance that yields an approximate Near Neighbor Search algorithm with the asymptotically optimal running time exponent. Unlike earlier algorithms with this property (e.g., Spherical LSH [1, 2]), our algorithm is also practical, improving upon the well-studied hyperplane LSH [3] in practice. We also introduce a multiprobe version of this algorithm and conduct an experimental evaluation on real and synthetic data sets. We complement the above positive results with a fine-grained lower bound for the quality of any LSH family for angular distance. Our lower bound implies that the above LSH family exhibits a trade-off between evaluation time and quality that is close to optimal for a natural class of LSH functions.

Retrieving the most similar vector embeddings to a given query among a massive collection of vectors has long been a key component of countless real-world applications. The recently introduced Retrieval-Augmented Generation is one of the most prominent examples. For many of these applications, the database evolves over time by inserting new data and removing outdated data. In these cases, the retrieval problem is known as streaming similarity search. While Locally-Adaptive Vector Quantization (LVQ), a highly efficient vector compression method, yields state-of-the-art search performance for non-evolving databases, its usefulness in the streaming setting has not been yet established. In this work, we study LVQ in streaming similarity search. In support of our evaluation, we introduce two improvements of LVQ: Turbo LVQ and multi-means LVQ that boost its search performance by up to 28% and 27%, respectively. Our studies show that LVQ and its new variants enable blazing fast vector search, outperforming its closest competitor by up to 9.4x for identically distributed data and by up to 8.8x under the challenging scenario of data distribution shifts (i.e., where the statistical distribution of the data changes over time). We release our contributions as part of Scalable Vector Search, an open-source library for high-performance similarity search.

In recent years, many hashing methods have been proposed for the cross-modal retrieval task. However, there are still some issues that need to be further explored. For example, some of them relax the binary constraints to generate the hash codes, which may generate large quantization error. Although some discrete schemes have been proposed, most of them are time-consuming. In addition, most of the existing supervised hashing methods use an n x n similarity matrix during the optimization, making them unscalable. To address these issues, in this paper, we present a novel supervised cross-modal hashing method—Scalable disCRete mATrix faCtorization Hashing, SCRATCH for short. It leverages the collective matrix factorization on the kernelized features and the semantic embedding with labels to find a latent semantic space to preserve the intra- and inter-modality similarities. In addition, it incorporates the label matrix instead of the similarity matrix into the loss function. Based on the proposed loss function and the iterative optimization algorithm, it can learn the hash functions and binary codes simultaneously. Moreover, the binary codes can be generated discretely, reducing the quantization error generated by the relaxation scheme. Its time complexity is linear to the size of the dataset, making it scalable to large-scale datasets. Extensive experiments on three benchmark datasets, namely, Wiki, MIRFlickr-25K, and NUS-WIDE, have verified that our proposed SCRATCH model outperforms several state-of-the-art unsupervised and supervised hashing methods for cross-modal retrieval.

Deep hashing techniques have emerged as the predominant approach for efficient image retrieval. Traditionally, these methods utilize pre-trained convolutional neural networks (CNNs) such as AlexNet and VGG-16 as feature extractors. However, the increasing complexity of datasets poses challenges for these backbone architectures in capturing meaningful features essential for effective image retrieval. In this study, we explore the efficacy of employing high-resolution features learned through state-of-the-art techniques for image retrieval tasks. Specifically, we propose a novel methodology that utilizes High-Resolution Networks (HRNets) as the backbone for the deep hashing task, termed High-Resolution Hashing Network (HHNet). Our approach demonstrates superior performance compared to existing methods across all tested benchmark datasets, including CIFAR-10, NUS-WIDE, MS COCO, and ImageNet. This performance improvement is more pronounced for complex datasets, which highlights the need to learn high-resolution features for intricate image retrieval tasks. Furthermore, we conduct a comprehensive analysis of different HRNet configurations and provide insights into the optimal architecture for the deep hashing task

Virtual screening (VS) is a critical step in computer-aided drug discovery, aiming to identify molecules that bind to a specific target receptor like protein. Traditional VS methods, such as docking, are often too time-consuming for screening large-scale molecular databases. Recent advances in deep learning have demonstrated that learning vector representations for both proteins and molecules using contrastive learning can outperform traditional docking methods. However, given that target databases often contain billions of molecules, real-valued vector representations adopted by existing methods can still incur significant memory and time costs in VS. To address this problem, in this paper we propose a hashing-based contrastive learning method, called DrugHash, for VS. DrugHash treats VS as a retrieval task that uses efficient binary hash codes for retrieval. In particular, DrugHash designs a simple yet effective hashing strategy to enable end-to-end learning of binary hash codes for both protein and molecule modalities, which can dramatically reduce the memory and time costs with higher accuracy compared with existing methods. Experimental results show that DrugHash can outperform existing methods to achieve state-of-the-art accuracy, with a memory saving of 32\(\times\) and a speed improvement of 3.5\(\times\).

Binary embedding of high-dimensional data requires long codes to preserve the discriminative power of the input space. Traditional binary coding methods often suffer from very high computation and storage costs in such a scenario. To address this problem, we propose Circulant Binary Embedding (CBE) which generates binary codes by projecting the data with a circulant matrix. The circulant structure enables the use of Fast Fourier Transformation to speed up the computation. Compared to methods that use unstructured matrices, the proposed method improves the time complexity from O(d^2 ) to O(d log d), and the space complexity from O(d^2) to O(d) where d is the input dimensionality. We also propose a novel time-frequency alternating optimization to learn data-dependent circulant projections, which alternatively minimizes the objective in original and Fourier domains. We show by extensive experiments that the proposed approach gives much better performance than the state-of-the-art approaches for fixed time, and provides much faster computation with no performance degradation for fixed number of bits.

Unsupervised cross-modal hashing (UCMH) has become a hot topic recently. Current UCMH focuses on exploring data similarities. However, current UCMH methods calculate the similarity between two data, mainly relying on the two data’s cross-modal features. These methods suffer from inaccurate similarity problems that result in a suboptimal retrieval Hamming space, because the cross-modal features between the data are not sufficient to describe the complex data relationships, such as situations where two data have different feature representations but share the inherent concepts. In this paper, we devise a deep graph-neighbor coherence preserving network (DGCPN). Specifically, DGCPN stems from graph models and explores graph-neighbor coherence by consolidating the information between data and their neighbors. DGCPN regulates comprehensive similarity preserving losses by exploiting three types of data similarities (i.e., the graph-neighbor coherence, the coexistent similarity, and the intra- and inter-modality consistency) and designs a half-real and half-binary optimization strategy to reduce the quantization errors during hashing. Essentially, DGCPN addresses the inaccurate similarity problem by exploring and exploiting the data’s intrinsic relationships in a graph. We conduct extensive experiments on three public UCMH datasets. The experimental results demonstrate the superiority of DGCPN, e.g., by improving the mean average precision from 0.722 to 0.751 on MIRFlickr-25K using 64-bit hashing codes to retrieval texts from images. We will release the source code package and the trained model on https://github.com/Atmegal/DGCPN.

Semantic hashing is a powerful paradigm for representing texts as compact binary hash codes. The explosion of short text data has spurred the demand of few-bits hashing. However, the performance of existing semantic hashing methods cannot be guaranteed when applied to few-bits hashing because of severe information loss. In this paper, we present a simple but effective unsupervised neural generative semantic hashing method with a focus on few-bits hashing. Our model is built upon variational autoencoder and represents each hash bit as a Bernoulli variable, which allows the model to be end-to-end trainable. To address the issue of information loss, we introduce a set of auxiliary implicit topic vectors. With the aid of these topic vectors, the generated hash codes are not only low-dimensional representations of the original texts but also capture their implicit topics. We conduct comprehensive experiments on four datasets. The results demonstrate that our approach achieves significant improvements over state-of-the-art semantic hashing methods in few-bits hashing.

Hashing techniques have recently been successfully applied to solve similarity search problems in the information retrieval field because of their significantly reduced storage and high-speed search capabilities. However, the hash codes learned from most recent cross-modal hashing methods lack the ability to comprehensively preserve adequate information, resulting in a less than desirable performance. To solve this limitation, we propose a novel method termed Nonlinear Robust Discrete Hashing (NRDH), for cross-modal retrieval. The main idea behind NRDH is motivated by the success of neural networks, i.e., nonlinear descriptors, in the field of representation learning, and the use of nonlinear descriptors instead of simple linear transformations is more in line with the complex relationships that exist between common latent representation and heterogeneous multimedia data in the real world. In NRDH, we first learn a common latent representation through nonlinear descriptors to encode complementary and consistent information from the features of the heterogeneous multimedia data. Moreover, an asymmetric learning scheme is proposed to correlate the learned hash codes with the common latent representation. Empirically, we demonstrate that NRDH is able to successfully generate a comprehensive common latent representation that significantly improves the quality of the learned hash codes. Then, NRDH adopts a linear learning strategy to fast learn the hash function with the learned hash codes. Extensive experiments performed on two benchmark datasets highlight the superiority of NRDH over several state-of-the-art methods.

Hash function learning has been widely used for largescale image retrieval because of the efficiency of computation and storage. We introduce AdaLabelHash, a binary hash function learning approach via deep neural networks in this paper. In AdaLabelHash, class label representations are variables that are adapted during the backward network training procedure. We express the labels as hypercube vertices in a K-dimensional space, and the class label representations together with the network weights are updated in the learning process. As the label representations (or referred to as codewords in this work), are learned from data, semantically similar classes will be assigned with the codewords that are close to each other in terms of Hamming distance in the label space. The codewords then serve as the desired output of the hash function learning, and yield compact and discriminating binary hash representations. AdaLabelHash is easy to implement, which can jointly learn label representations and infer compact binary codes from data. It is applicable to both supervised and semi-supervised hash. Experimental results on standard benchmarks demonstrate the satisfactory performance of AdaLabelHash.

Due to the high storage and search efficiency, hashing has become prevalent for large-scale similarity search. Particularly, deep hashing methods have greatly improved the search performance under supervised scenarios. In contrast, unsupervised deep hashing models can hardly achieve satisfactory performance due to the lack of reliable supervisory similarity signals. To address this issue, we propose a novel deep unsupervised hashing model, dubbed DistillHash, which can learn a distilled data set consisted of data pairs, which have confidence similarity signals. Specifically, we investigate the relationship between the initial noisy similarity signals learned from local structures and the semantic similarity labels assigned by a Bayes optimal classifier. We show that under a mild assumption, some data pairs, of which labels are consistent with those assigned by the Bayes optimal classifier, can be potentially distilled. Inspired by this fact, we design a simple yet effective strategy to distill data pairs automatically and further adopt a Bayesian learning framework to learn hash functions from the distilled data set. Extensive experimental results on three widely used benchmark datasets show that the proposed DistillHash consistently accomplishes the stateof-the-art search performance.

The evolution of 3D generative modeling has been notably propelled by the adoption of 2D diffusion models. Despite this progress, the cumbersome optimization process per se presents a critical hurdle to efficiency. In this paper, we introduce Hash3D, a universal acceleration for 3D generation without model training. Central to Hash3D is the insight that feature-map redundancy is prevalent in images rendered from camera positions and diffusion time-steps in close proximity. By effectively hashing and reusing these feature maps across neighboring timesteps and camera angles, Hash3D substantially prevents redundant calculations, thus accelerating the diffusion model’s inference in 3D generation tasks. We achieve this through an adaptive grid-based hashing. Surprisingly, this feature-sharing mechanism not only speed up the generation but also enhances the smoothness and view consistency of the synthesized 3D objects. Our experiments covering 5 text-to-3D and 3 image-to-3D models, demonstrate Hash3D’s versatility to speed up optimization, enhancing efficiency by 1.3 to 4 times. Additionally, Hash3D’s integration with 3D Gaussian splatting largely speeds up 3D model creation, reducing text-to-3D processing to about 10 minutes and image-to-3D conversion to roughly 30 seconds. The project page is at https://adamdad.github.io/hash3D/.

Existing data-dependent hashing methods usually learn hash functions from the pairwise or triplet data relationships, which only capture the data similarity locally, and often suffer low learning efficiency and low collision rate. In this work, we propose a new global similarity metric, termed as central similarity, with which the hash codes for similar data pairs are encouraged to approach a common center and those for dissimilar pairs to converge to different centers, to improve hash learning efficiency and retrieval accuracy. We principally formulate the computation of the proposed central similarity metric by introducing a new concept, i.e. hash center that refers to a set of data points scattered in the Hamming space with sufficient mutual distance between each other. We then provide an efficient method to construct well separated hash centers by leveraging the Hadamard matrix and Bernoulli distributions. Finally, we propose the Central Similarity Hashing (CSH) that optimizes the central similarity between data points w.r.t. their hash centers instead of optimizing the local similarity. The CSH is generic and applicable to both image and video hashing. Extensive experiments on large-scale image and video retrieval demonstrate CSH can generate cohesive hash codes for similar data pairs and dispersed hash codes for dissimilar pairs, and achieve noticeable boost in retrieval performance, i.e. 3%-20% in mAP over the previous state-of-the-art. The codes are in: https://github.com/yuanli2333/ Hadamard-Matrix-for-hashing

This paper tackles a rarely explored but critical problem within learning to hash, i.e., to learn hash codes that effectively discriminate hard similar and dissimilar examples, to empower large-scale image retrieval. Hard similar examples refer to image pairs from the same semantic class that demonstrate some shared appearance but have different fine-grained appearance. Hard dissimilar examples are image pairs that come from different semantic classes but exhibit similar appearance. These hard examples generally have a small distance due to the shared appearance. Therefore, effective encoding of the hard examples can well discriminate the relevant images within a small Hamming distance, enabling more accurate retrieval in the top-ranked returned images. However, most existing hashing methods cannot capture this key information as their optimization is dominated byeasy examples, i.e., distant similar/dissimilar pairs that share no or limited appearance. To address this problem, we introduce a novel Gamma distribution-enabled and symmetric Kullback-Leibler divergence-based loss, which is dubbed dual hinge loss because it works similarly as imposing two smoothed hinge losses on the respective similar and dissimilar pairs. Specifically, the loss enforces exponentially variant penalization on the hard similar (dissimilar) examples to emphasize and learn their fine-grained difference. It meanwhile imposes a bounding penalization on easy similar (dissimilar) examples to prevent the dominance of the easy examples in the optimization while preserving the high-level similarity (dissimilarity). This enables our model to well encode the key information carried by both easy and hard examples. Extensive empirical results on three widely-used image retrieval datasets show that (i) our method consistently and substantially outperforms state-of-the-art competing methods using hash codes of the same length and (ii) our method can use significantly (e.g., 50%-75%) shorter hash codes to perform substantially better than, or comparably well to, the competing methods.

Yandex DEEP-1B image descriptor dataset consisting of the projected and normalized outputs from the last fully-connected layer of the GoogLeNet model, which was pretrained on the Imagenet classification task.

Hashing, as a popular approximate nearest neighbor search, has been widely used for large-scale similarity search. Recently, a spectrum of machine learning methods are utilized to learn similarity-preserving binary codes. However, most of them directly encode the explicit features, keywords, which fail to preserve the accurate semantic similarities in binary code beyond keyword matching, especially on short texts. Here we propose a novel text hashing framework with convolutional neural networks. In particular, we first embed the keyword features into compact binary code with a locality preserving constraint. Meanwhile word features and position features are together fed into a convolutional network to learn the implicit features which are further incorporated with the explicit features to fit the pre-trained binary code. Such base method can be successfully accomplished without any external tags/labels, and other three model variations are designed to integrate tags/labels. Experimental results show the superiority of our proposed approach over several state-of-the-art hashing methods when tested on one short text dataset as well as one normal text dataset.

Large-scale data mining and retrieval applications have increasingly turned to compact binary data representations as a way to achieve both fast queries and efficient data storage; many algorithms have been proposed for learning effective binary encodings. Most of these algorithms focus on learning a set of projection hyperplanes for the data and simply binarizing the result from each hyperplane, but this neglects the fact that informativeness may not be uniformly distributed across the projections. In this paper, we address this issue by proposing a novel adaptive quantization (AQ) strategy that adaptively assigns varying numbers of bits to different hyperplanes based on their information content. Our method provides an information-based schema that preserves the neighborhood structure of data points, and we jointly find the globally optimal bit-allocation for all hyperplanes. In our experiments, we compare with state-of-the-art methods on four large-scale datasets and find that our adaptive quantization approach significantly improves on traditional hashing methods.

Hashing-based fast nearest neighbor search technique has attracted great attention in both research and industry areas recently. Many existing hashing approaches encode data with projection-based hash functions and represent each projected dimension by 1-bit. However, the dimensions with high variance hold large energy or information of data but treated equivalently as dimensions with low variance, which leads to a serious information loss. In this paper, we introduce a novel hashing algorithm called Harmonious Hashing which aims at learning hash functions with low information loss. Specifically, we learn a set of optimized projections to preserve the maximum cumulative energy and meet the constraint of equivalent variance on each dimension as much as possible. In this way, we could minimize the information loss after binarization. Despite the extreme simplicity, our method outperforms superiorly to many state-of-the-art hashing methods in large-scale and high-dimensional nearest neighbor search experiments.

Hashing is a popular approximate nearest neighbor search approach for large-scale image retrieval. Supervised hashing, which incorporates similarity/dissimilarity information on entity pairs to improve the quality of hashing function learning, has recently received increasing attention. However, in the existing supervised hashing methods for images, an input image is usually encoded by a vector of hand-crafted visual features. Such hand-crafted feature vectors do not necessarily preserve the accurate semantic similarities of images pairs, which may often degrade the performance of hashing function learning. In this paper, we propose a supervised hashing method for image retrieval, in which we automatically learn a good image representation tailored to hashing as well as a set of hash functions. The proposed method has two stages. In the first stage, given the pairwise similarity matrix S over training images, we propose a scalable coordinate descent method to decompose S into a product of HHT where H is a matrix with each of its rows being the approximate hash code associated to a training image. In the second stage, we propose to simultaneously learn a good feature representation for the input images as well as a set of hash functions, via a deep convolutional network tailored to the learned hash codes in H and optionally the discrete class labels of the images. Extensive empirical evaluations on three benchmark datasets with different kinds of images show that the proposed method has superior performance gains over several state-of-the-art supervised and unsupervised hashing methods.

Hashing has shown great potential in large-scale image retrieval due to its storage and computation efficiency, especially the recent deep supervised hashing methods. To achieve promising performance, deep supervised hashing methods require a large amount of training data from different classes. However, when images of new categories emerge, existing deep hashing methods have to retrain the CNN model and generate hash codes for all the database images again, which is impractical for large-scale retrieval system. In this paper, we propose a novel deep hashing framework, called Deep Incremental Hashing Network (DIHN), for learning hash codes in an incremental manner. DIHN learns the hash codes for the new coming images directly, while keeping the old ones unchanged. Simultaneously, a deep hash function for query set is learned by preserving the similarities between training points. Extensive experiments on two widely used image retrieval benchmarks demonstrate that the proposed DIHN framework can significantly decrease the training time while keeping the state-of-the-art retrieval accuracy.

Bipartite graph hashing (BGH) is extensively used for Top-K search in Hamming space at low storage and inference costs. Recent research adopts graph convolutional hashing for BGH and has achieved the state-of-the-art performance. However, the contributions of its various influencing factors to hashing performance have not been explored in-depth, including the same/different sign count between two binary embeddings during Hamming space search (sign property), the contribution of sub-embeddings at each layer (model property), the contribution of different node types in the bipartite graph (node property), and the combination of augmentation methods. In this work, we build a lightweight graph convolutional hashing model named LightGCH by mainly removing the augmentation methods of the state-of-the-art model BGCH. By analyzing the contributions of each layer and node type to performance, as well as analyzing the Hamming similarity statistics at each layer, we find that the actual neighbors in the bipartite graph tend to have low Hamming similarity at the shallow layer, and all nodes tend to have high Hamming similarity at the deep layers in LightGCH. To tackle these problems, we propose a novel sign-guided framework SGBGH to make improvement, which uses sign-guided negative sampling to improve the Hamming similarity of neighbors, and uses sign-aware contrastive learning to help nodes learn more uniform representations. Experimental results show that SGBGH outperforms BGCH and LightGCH significantly in embedding quality.

Online hashing methods are efficient in learning the hash functions from the streaming data. However, when the hash functions change, the binary codes for the database have to be recomputed to guarantee the retrieval accuracy. Recomputing the binary codes by accumulating the whole database brings a timeliness challenge to the online retrieval process. In this paper, we propose a novel online hashing framework to update the binary codes efficiently without accumulating the whole database. In our framework, the hash functions are fixed and the projection functions are introduced to learn online from the streaming data. Therefore, inefficient updating of the binary codes by accumulating the whole database can be transformed to efficient updating of the binary codes by projecting the binary codes into another binary space. The queries and the binary code database are projected asymmetrically to further improve the retrieval accuracy. The experiments on two multi-label image databases demonstrate the effectiveness and the efficiency of our method for multi-label image retrieval.

In this paper, we propose a simple yet effective relaxation free method to learn more effective binary codes via policy gradient for scalable image search. While a variety of deep hashing methods have been proposed in recent years, most of them are confronted by the dilemma to obtain optimal binary codes in a truly end-to-end manner with nonsmooth sign activations. Unlike existing methods which usually employ a general relaxation framework to adapt to the gradient-based algorithms, our approach formulates the non-smooth part of the hashing network as sampling with a stochastic policy, so that the retrieval performance degradation caused by the relaxation can be avoided. Specifically, our method directly generates the binary codes and maximizes the expectation of rewards for similarity preservation, where the network can be trained directly via policy gradient. Hence, the differentiation challenge for discrete optimization can be naturally addressed, which leads to effective gradients and binary codes. Extensive experimental results on three benchmark datasets validate the effectiveness of the proposed method.

Our work focuses on tackling large-scale fine-grained image retrieval as ranking the images depicting the concept of interests (i.e., the same sub-category labels) highest based on the fine-grained details in the query. It is desirable to alleviate the challenges of both fine-grained nature of small inter-class variations with large intra-class variations and explosive growth of fine-grained data for such a practical task. In this paper, we propose an Attribute-Aware hashing Network (A-Net) for generating attribute-aware hash codes to not only make the retrieval process efficient, but also establish explicit correspondences between hash codes and visual attributes. Specifically, based on the captured visual representations by attention, we develop an encoder-decoder structure network of a reconstruction task to unsupervisedly distill high-level attribute-specific vectors from the appearance-specific visual representations without attribute annotations. A-Net is also equipped with a feature decorrelation constraint upon these attribute vectors to enhance their representation abilities. Finally, the required hash codes are generated by the attribute vectors driven by preserving original similarities. Qualitative experiments on five benchmark fine-grained datasets show our superiority over competing methods. More importantly, quantitative results demonstrate the obtained hash codes can strongly correspond to certain kinds of crucial properties of fine-grained objects.

Implementing cross-modal hashing between 2D images and 3D point-cloud data is a growing concern in real-world retrieval systems. Simply applying existing cross-modal approaches to this new task fails to adequately capture latent multi-modal semantics and effectively bridge the modality gap between 2D and 3D. To address these issues without relying on hand-crafted labels, we propose contrastive masked autoencoders based self-supervised hashing (CMAH) for retrieval between images and point-cloud data. We start by contrasting 2D-3D pairs and explicitly constraining them into a joint Hamming space. This contrastive learning process ensures robust discriminability for the generated hash codes and effectively reduces the modality gap. Moreover, we utilize multi-modal auto-encoders to enhance the model’s understanding of multi-modal semantics. By completing the masked image/point-cloud data modeling task, the model is encouraged to capture more localized clues. In addition, the proposed multi-modal fusion block facilitates fine-grained interactions among different modalities. Extensive experiments on three public datasets demonstrate that the proposed CMAH significantly outperforms all baseline methods.

Learning to hash has become popular for video retrieval due to its fast speed and low storage consumption. Previous efforts formulate video hashing as training a binary auto-encoder, for which noncontinuous latent representations are optimized by the biased straight-through (ST) back-propagation heuristic. We propose to formulate video hashing as learning a discrete variational auto-encoder with the factorized Bernoulli latent distribution, termed as Bernoulli variational auto-encoder (BerVAE). The corresponding evidence lower bound (ELBO) in our BerVAE implementation leads to closed-form gradient expression, which can be applied to achieve principled training along with some other unbiased gradient estimators. BerVAE enables uncertainty-aware video hashing by predicting the probability distribution of video hash code-words, thus providing reliable uncertainty quantification. Experiments on both simulated and real-world large-scale video data demonstrate that our BerVAE trained with unbiased gradient estimators can achieve the state-of-the-art retrieval performance. Furthermore, we show that quantified uncertainty is highly correlated to video retrieval performance, which can be leveraged to further improve the retrieval accuracy. Our code is available at https://github.com/wangyucheng1234/BerVAE

Hashing based Approximate Nearest Neighbor (ANN) search has attracted much attention due to its fast query time and drastically reduced storage. However, most of the hashing methods either use random projections or extract principal directions from the data to derive hash functions. The resulting embedding suffers from poor discrimination when compact codes are used. In this paper, we propose a novel data-dependent projection learning method such that each hash function is designed to correct the errors made by the previous one sequentially. The proposed method easily adapts to both unsupervised and semi-supervised scenarios and shows significant performance gains over the state-ofthe-art methods on two large datasets containing up to 1 million points.

The problem of cross-modal similarity search, which aims at making efficient and accurate queries across multiple domains, has become a significant and important research topic. Composite quantization, a compact coding solution superior to hashing techniques, has shown its effectiveness for similarity search. However, most existing works utilizing composite quantization to search multi-domain content only consider either pairwise similarity information or class label information across different domains, which fails to tackle the semi-supervised problem in composite quantization. In this paper, we address the semi-supervised quantization problem by considering: (i) pairwise similarity information (without class label information) across different domains, which captures the intra-document relation, (ii) cross-domain data with class label which can help capture inter-document relation, and (iii) cross-domain data with neither pairwise similarity nor class label which enables the full use of abundant unlabelled information. To the best of our knowledge, we are the first to consider both supervised information (pairwise similarity + class label) and unsupervised information (neither pairwise similarity nor class label) simultaneously in composite quantization. A challenging problem arises: how can we jointly handle these three sorts of information across multiple domains in an efficient way? To tackle this challenge, we propose a novel semi-supervised deep quantization (SSDQ) model that takes both supervised and unsupervised information into account. The proposed SSDQ model is capable of incorporating the above three kinds of information into one single framework when utilizing composite quantization for accurate and efficient queries across different domains. More specifically, we employ a modified deep autoencoder for better latent representation and formulate pairwise similarity loss, supervised quantization loss as well as unsupervised distribution match loss to handle all three types of information. The extensive experiments demonstrate the significant improvement of SSDQ over several state-of-the-art methods on various datasets.

Nearest neighbor search is a problem of finding the data points from the database such that the distances from them to the query point are the smallest. Learning to hash is one of the major solutions to this problem and has been widely studied recently. In this paper, we present a comprehensive survey of the learning to hash algorithms, categorize them according to the manners of preserving the similarities into: pairwise similarity preserving, multiwise similarity preserving, implicit similarity preserving, as well as quantization, and discuss their relations. We separate quantization from pairwise similarity preserving as the objective function is very different though quantization, as we show, can be derived from preserving the pairwise similarities. In addition, we present the evaluation protocols, and the general performance analysis, and point out that the quantization algori

Deep quantization methods have shown high efficiency on large-scale image retrieval. However, current models heavily rely on ground-truth information, hindering the application of quantization in label-hungry scenarios. A more realistic demand is to learn from inexhaustible uploaded images that are associated with informal tags provided by amateur users. Though such sketchy tags do not obviously reveal the labels, they actually contain useful semantic information for supervising deep quantization. To this end, we propose Weakly-Supervised Deep Hyperspherical Quantization (WSDHQ), which is the first work to learn deep quantization from weakly tagged images. Specifically, 1) we use word embeddings to represent the tags and enhance their semantic information based on a tag correlation graph. 2) To better preserve semantic information in quantization codes and reduce quantization error, we jointly learn semantics-preserving embeddings and supervised quantizer on hypersphere by employing a well-designed fusion layer and tailor-made loss functions. Extensive experiments show that WSDHQ can achieve state-of-art performance on weakly-supervised compact coding. Code is available at https://github.com/gimpong/AAAI21-WSDHQ.

Due to its powerful capability of representation learning and high-efficiency computation, deep hashing has made significant progress in large-scale image retrieval. However, deep hashing networks are vulnerable to adversarial examples, which is a practical secure problem but seldom studied in hashing-based retrieval field. In this paper, we propose a novel prototype-supervised adversarial network (ProS-GAN), which formulates a flexible generative architecture for efficient and effective targeted hashing attack. To the best of our knowledge, this is the first generation-based method to attack deep hashing networks. Generally, our proposed framework consists of three parts, i.e., a PrototypeNet, a generator and a discriminator. Specifically, the designed PrototypeNet embeds the target label into the semantic representation and learns the prototype code as the category-level representative of the target label. Moreover, the semantic representation and the original image are jointly fed into the generator for flexible targeted attack. Particularly, the prototype code is adopted to supervise the generator to construct the targeted adversarial example by minimizing the Hamming distance between the hash code of the adversarial example and the prototype code. Furthermore, the generator is against the discriminator to simultaneously encourage the adversarial examples visually realistic and the semantic representation informative. Extensive experiments verify that the proposed framework can efficiently produce adversarial examples with better targeted attack performance and transferability over state-of-the-art targeted attack methods of deep hashing.

Cross-modal hashing has been widely investigated recently for its efficiency in large-scale cross-media retrieval. However, most existing cross-modal hashing methods learn hash functions in a batch-based learning mode. Such mode is not suitable for large-scale data sets due to the large memory consumption and loses its efficiency when training streaming data. Online cross-modal hashing can deal with the above problems by learning hash model in an online learning process. However, existing online cross-modal hashing methods cannot update hash codes of old data by the newly learned model. In this paper, we propose Online Collective Matrix Factorization Hashing (OCMFH) based on collective matrix factorization hashing (CMFH), which can adaptively update hash codes of old data according to dynamic changes of hash model without accessing to old data. Specifically, it learns discriminative hash codes for streaming data by collective matrix factorization in an online optimization scheme. Unlike conventional CMFH which needs to load the entire data points into memory, the proposed OCMFH retrains hash functions only by newly arriving data points. Meanwhile, it generates hash codes of new data and updates hash codes of old data by the latest updated hash model. In such way, hash codes of new data and old data are well-matched. Furthermore, a zero mean strategy is developed to solve the mean-varying problem in the online hash learning process. Extensive experiments on three benchmark data sets demonstrate the effectiveness and efficiency of OCMFH on online cross-media retrieval.

Known for efficient computation and easy storage, hashing has been extensively explored in cross-modal retrieval. The majority of current hashing models are predicated on the premise of a direct one-to-one mapping between data points. However, in real practice, data correspondence across modalities may be partially provided. In this research, we introduce an innovative unsupervised hashing technique designed for semi-paired cross-modal retrieval tasks, named Reconstruction Relations Embedded Hashing (RREH). RREH assumes that multi-modal data share a common subspace. For paired data, RREH explores the latent consistent information of heterogeneous modalities by seeking a shared representation. For unpaired data, to effectively capture the latent discriminative features, the high-order relationships between unpaired data and anchors are embedded into the latent subspace, which are computed by efficient linear reconstruction. The anchors are sampled from paired data, which improves the efficiency of hash learning. The RREH trains the underlying features and the binary encodings in a unified framework with high-order reconstruction relations preserved. With the well devised objective function and discrete optimization algorithm, RREH is designed to be scalable, making it suitable for large-scale datasets and facilitating efficient cross-modal retrieval. In the evaluation process, the proposed is tested with partially paired data to establish its superiority over several existing methods.

In this paper, we investigate the problem of unsupervised domain adaptive hashing, which leverage knowledge from a label-rich source domain to expedite learning to hash on a label-scarce target domain. Although numerous existing approaches attempt to incorporate transfer learning techniques into deep hashing frameworks, they often neglect the essential invariance for adequate alignment between these two domains. Worse yet, these methods fail to distinguish between causal and non-causal effects embedded in images, rendering cross-domain retrieval ineffective. To address these challenges, we propose an Invariance-acquired Domain AdaptivE HAshing (IDEA) model. Our IDEA first decomposes each image into a causal feature representing label information, and a non-causal feature indicating domain information. Subsequently, we generate discriminative hash codes using causal features with consistency learning on both source and target domains. More importantly, we employ a generative model for synthetic samples to simulate the intervention of various non-causal effects, ultimately minimizing their impact on hash codes for domain invariance. Comprehensive experiments conducted on benchmark datasets validate the superior performance of our IDEA compared to a variety of competitive baselines.

Hashing is one of the effective techniques for fast Approximate Nearest Neighbour (ANN) search. Traditional single-bit quantization (SBQ) in most hashing methods incurs lots of quantization error which seriously degrades the search performance. To address the limitation of SBQ, researchers have proposed promising multi-bit quantization (MBQ) methods to quantize each projection dimension with multiple bits. However, some MBQ methods need to adopt specific distance for binary code matching instead of the original Hamming distance, which would significantly decrease the retrieval speed. Two typical MBQ methods Hierarchical Quantization and Double Bit Quantization retain the Hamming distance, but both of them only consider the projection dimensions during quantization, ignoring the neighborhood structure of raw data inherent in Euclidean space. In this paper, we propose a multi-bit quantization method named Hamming Compatible Quantization (HCQ) to preserve the capability of similarity metric between Euclidean space and Hamming space by utilizing the neighborhood structure of raw data. Extensive experiment results have shown our approach significantly improves the performance of various stateof-the-art hashing methods while maintaining fast retrieval speed.

Existing deep hashing approaches fail to fully explore semantic correlations and neglect the effect of linguistic context on visual attention learning, leading to inferior performance. This paper proposes a dual-stream learning framework, dubbed Deep Collaborative Discrete Hashing (DCDH), which constructs a discriminative common discrete space by collaboratively incorporating the shared and individual semantics deduced from visual features and semantic labels. Specifically, the context-aware representations are generated by employing the outer product of visual embeddings and semantic encodings. Moreover, we reconstruct the labels and introduce the focal loss to take advantage of frequent and rare concepts. The common binary code space is built on the joint learning of the visual representations attended by language, the semantic-invariant structure construction and the label distribution correction. Extensive experiments demonstrate the superiority of our method.

To quantify different types of uncertainty when deriving hash-codes for image retrieval, we develop a probabilistic hashing model(ProbHash). Sampling-based hypothesis testing is then derived for hashing with uncertainty quantification(HashUQ) in ProbHash to improve the granularity of hashing-based retrieval by prioritizing the data with confident hash-codes. HashUQ can drastically improve the retrieval performance without sacrificing computational efficiency. For efficient deployment of HashUQ in real-world applications, we discretize the quantified uncertainty to reduce the potential storage overhead. Experimental results show that our HashUQ can achieve state-of-the-art retrieval performance on three image datasets. Ablation experiments on model hyperparameters, different model components, and effects of UQ are also provided with performance comparisons. Our code is available at https://github.com/QianLab/HashUQ.

Locality-sensitive hashing (LSH) is a fundamental algorithmic technique widely employed in large-scale data processing applications, such as nearest-neighbor search, entity resolution, and clustering. However, its applicability in some real-world scenarios is limited due to the need for careful design of hashing functions that align with specific metrics. Existing LSH-based Entity Blocking solutions primarily rely on generic similarity metrics such as Jaccard similarity, whereas practical use cases often demand complex and customized similarity rules surpassing the capabilities of generic similarity metrics. Consequently, designing LSH functions for these customized similarity rules presents considerable challenges. In this research, we propose a neuralization approach to enhance locality-sensitive hashing by training deep neural networks to serve as hashing functions for complex metrics. We assess the effectiveness of this approach within the context of the entity resolution problem, which frequently involves the use of task-specific metrics in real-world applications. Specifically, we introduce NLSHBlock (Neural-LSH Block), a novel blocking methodology that leverages pre-trained language models, fine-tuned with a novel LSH-based loss function. Through extensive evaluations conducted on a diverse range of real-world datasets, we demonstrate the superiority of NLSHBlock over existing methods, exhibiting significant performance improvements. Furthermore, we showcase the efficacy of NLSHBlock in enhancing the performance of the entity matching phase, particularly within the semi-supervised setting.

Multimodal hashing is essential to cross-media similarity search for its low storage cost and fast query speed. Most existing multimodal hashing methods embedded heterogeneous data into a common low-dimensional Hamming space, and then rounded the continuous embeddings to obtain the binary codes. Yet they usually neglect the inherent discrete nature of hashing for relaxing the discrete constraints, which will cause degraded retrieval performance especially for long codes. For this purpose, a novel Semantic Topic Multimodal Hashing (STMH) is developed by considering latent semantic information in coding procedure. It first discovers clustering patterns of texts and robust factorizes the matrix of images to obtain multiple semantic topics of texts and concepts of images. Then the learned multimodal semantic features are transformed into a common subspace by their correlations. Finally, each bit of unified hash code can be generated directly by figuring out whether a topic or concept is contained in a text or an image. Therefore, the obtained model by STMH is more suitable for hashing scheme as it directly learns discrete hash codes in the coding process. Experimental results demonstrate that the proposed method outperforms several state-of-the-art methods.

en a surge of interest in methods based on “semantic hashing”, i.e. compact binary codes of data-points so that the Hamming distance between codewords correlates with similarity. In reviewing and comparing existing methods, we show that their relative performance can change drastically depending on the definition of ground-truth neighbors. Motivated by this finding, we propose a new formulation for learning binary codes which seeks to reconstruct the affinity between datapoints, rather than their distances. We show that this criterion is intractable to solve exactly, but a spectral relaxation gives an algorithm where the bits correspond to thresholded eigenvectors of the affinity matrix, and as the number of datapoints goes to infinity these eigenvectors converge to eigenfunctions of Laplace-Beltrami operators, similar to the recently proposed Spectral Hashing (SH) method. Unlike SH whose performance may degrade as the number of bits increases, the optimal code using our formulation is guaranteed to faithfully reproduce the affinities as the number of bits increases. We show that the number of eigenfunctions needed may increase exponentially with dimension, but introduce a “kernel trick” to allow us to compute with an exponentially large number of bits but using only memory and computation that grows linearly with dimension. Experiments shows that MDSH outperforms the state-of-the art, especially in the challenging regime of small distance thresholds.

Hashing techniques are powerful for approximate nearest neighbour (ANN) search. Existing quantization methods in hashing are all focused on scalar quantization (SQ) which is inferior in utilizing the inherent data distribution. In this paper, we propose a novel vector quantization (VQ) method named affinity preserving quantization (APQ) to improve the quantization quality of projection values, which has significantly boosted the performance of state-of-the-art hashing techniques. In particular, our method incorporates the neighbourhood structure in the pre- and post-projection data space into vector quantization. APQ minimizes the quantization errors of projection values as well as the loss of affinity property of original space. An effective algorithm has been proposed to solve the joint optimization problem in APQ, and the extension to larger binary codes has been resolved by applying product quantization to APQ. Extensive experiments have shown that APQ consistently outperforms the state-of-the-art quantization methods, and has significantly improved the performance of various hashing techniques.

Recently, binary representation has been proposed as a novel representation that lies between continuous and discrete representations. It exhibits considerable information-preserving capability when being used to replace continuous input vectors. In this paper, we investigate the feasibility of further introducing it to the output side, aiming to allow models to output binary labels instead. To preserve the structural information on the output side along with label information, we extend the previous contrastive hashing method as structured contrastive hashing. More specifically, we upgrade CKY from label-level to bit-level, define a new similarity function with span marginal probabilities, and introduce a novel contrastive loss function with a carefully designed instance selection strategy. Our model achieves competitive performance on various structured prediction tasks, and demonstrates that binary representation can be considered a novel representation that further bridges the gap between the continuous nature of deep learning and the discrete intrinsic property of natural languages.

Semantic hashing seeks compact binary codes of data-points so that the Hamming distance between codewords correlates with semantic similarity. In this paper, we show that the problem of finding a best code for a given dataset is closely related to the problem of graph partitioning and can be shown to be NP hard. By relaxing the original problem, we obtain a spectral method whose solutions are simply a subset of thresholded eigenvectors of the graph Laplacian. By utilizing recent results on convergence of graph Laplacian eigenvectors to the Laplace-Beltrami eigenfunctions of manifolds, we show how to efficiently calculate the code of a novel datapoint. Taken together, both learning the code and applying it to a novel point are extremely simple. Our experiments show that our codes outperform the state-of-the art.

Locality sensitive hashing (LSH) is a fundamental algorithmic toolkit used by data scientists for approximate nearest neighbour search problems that have been used extensively in many large scale data processing applications such as near duplicate detection, nearest neighbour search, clustering, etc. In this work, we aim to propose faster and space efficient locality sensitive hash functions for Euclidean distance and cosine similarity for tensor data. Typically, the naive approach for obtaining LSH for tensor data involves first reshaping the tensor into vectors, followed by applying existing LSH methods for vector data \(E2LSH\) and \(SRP\). However, this approach becomes impractical for higher order tensors because the size of the reshaped vector becomes exponential in the order of the tensor. Consequently, the size of LSH parameters increases exponentially. To address this problem, we suggest two methods for LSH for Euclidean distance and cosine similarity, namely \(CP-E2LSH\), \(TT-E2LSH\), and \(CP-SRP\), \(TT-SRP\), respectively, building on \(CP\) and tensor train \((TT)\) decompositions techniques. Our approaches are space efficient and can be efficiently applied to low rank \(CP\) or \(TT\) tensors. We provide a rigorous theoretical analysis of our proposal on their correctness and efficacy.

By quantizing network weights and activations to low bitwidth, we can obtain hardware-friendly and energy-efficient networks. However, existing quantization techniques utilizing the straight-through estimator and piecewise constant functions face the issue of how to represent originally high-bit input data with low-bit values. To fully quantize deep neural networks, we propose pixel embedding, which replaces each float-valued input pixel with a vector of quantized values by using a lookup table. The lookup table or low-bit representation of pixels is differentiable and trainable by backpropagation. Such replacement of inputs with vectors is similar to word embedding in the natural language processing field. Experiments on ImageNet and CIFAR-100 show that pixel embedding reduces the top-5 error gap caused by quantizing the floating points at the first layer to only 1% for the ImageNet dataset, and the top-1 error gap caused by quantizing first and last layers to slightly over 1% for the CIFAR-100 dataset. The usefulness of pixel embedding is further demonstrated by inference time measurements, which demonstrate over 1.7 times speedup compared to floating point precision first layer.

With the advent of the Internet, billions of images are now freely available online and constitute a dense sampling of the visual world. Using a variety of non-parametric methods, we explore this world with the aid of a large dataset of 79,302,017 images collected from the Web. Motivated by psychophysical results showing the remarkable tolerance of the human visual system to degradations in image resolution, the images in the dataset are stored as 32 × 32 color images. Each image is loosely labeled with one of the 75,062 non-abstract nouns in English, as listed in the Wordnet lexical database. Hence the image database gives a comprehensive coverage of all object categories and scenes. The semantic information from Wordnet can be used in conjunction with nearest-neighbor methods to perform object classification over a range of semantic levels minimizing the effects of labeling noise. For certain classes that are particularly prevalent in the dataset, such as people, we are able to demonstrate a recognition performance comparable to class-specific Viola-Jones style detectors.

We present an efficient and fast algorithm for computing approximate nearest neighbor fields between two images. Our method builds on the concept of Coherency-Sensitive Hashing (CSH), but uses a recent hashing scheme, Spherical Hashing (SpH), which is known to be better adapted to the nearest-neighbor problem for natural images. Cascaded Spherical Hashing concatenates different configurations of SpH to build larger Hash Tables with less elements in each bin to achieve higher selectivity. Our method amply outperforms existing techniques like PatchMatch and CSH, and the experimental results show that our algorithm is faster and more accurate than existing methods.

In approximate nearest neighbor search (ANNS) methods based on approximate proximity graphs, DiskANN achieves good recall-speed balance for large-scale datasets using both of RAM and storage. Despite it claims to save memory usage by loading compressed vectors by product quantization (PQ), its memory usage increases in proportion to the scale of datasets. In this paper, we propose All-in-Storage ANNS with Product Quantization (AiSAQ), which offloads the compressed vectors to storage. Our method achieves \(\sim\)10 MB memory usage in query search even with billion-scale datasets with minor performance degradation. AiSAQ also reduces the index load time before query search, which enables the index switch between muitiple billion-scale datasets and significantly enhances the flexibility of retrieval-augmented generation (RAG). This method is applicable to all graph-based ANNS algorithms and can be combined with higher-spec ANNS methods in the future.

Finding nearest neighbors has become an important operation on databases, with applications to text search, multimedia indexing, and many other areas. One popular algorithm for similarity search, especially for high dimensional data (where spatial indexes like kdtrees do not perform well) is Locality Sensitive Hashing (LSH), an approximation algorithm for finding similar objects. In this paper, we describe a new variant of LSH, called Parallel LSH (PLSH) designed to be extremely efficient, capable of scaling out on multiple nodes and multiple cores, and which supports highthroughput streaming of new data. Our approach employs several novel ideas, including: cache-conscious hash table layout, using a 2-level merge algorithm for hash table construction; an efficient algorithm for duplicate elimination during hash-table querying; an insert-optimized hash table structure and efficient data expiration algorithm for streaming data; and a performance model that accurately estimates performance of the algorithm and can be used to optimize parameter settings. We show that on a workload where we perform similarity search on a dataset of > 1 Billion tweets, with hundreds of millions of new tweets per day, we can achieve query times of 1–2.5 ms. We show that this is an order of magnitude faster than existing indexing schemes, such as inverted indexes. To the best of our knowledge, this is the fastest implementation of LSH, with table construction times up to 3.7x faster and query times that are 8.3x faster than a basic implementation.

Recently, due to the unprecedented growth of multimedia data, cross-modal hashing has gained increasing attention for the efficient cross-media retrieval. Typically, existing methods on crossmodal hashing treat labels of one instance independently but overlook the correlations among labels. Indeed, in many real-world scenarios, like the online fashion domain, instances (items) are labeled with a set of categories correlated by certain hierarchy. In this paper, we propose a new end-to-end solution for supervised cross-modal hashing, named HiCHNet, which explicitly exploits the hierarchical labels of instances. In particular, by the pre-established label hierarchy, we comprehensively characterize each modality of the instance with a set of layer-wise hash representations. In essence, hash codes are encouraged to not only preserve the layerwise semantic similarities encoded by the label hierarchy, but also retain the hierarchical discriminative capabilities. Due to the lack of benchmark datasets, apart from adapting the existing dataset FashionVC from fashion domain, we create a dataset from the online fashion platform Ssense consisting of 15, 696 image-text pairs labeled by 32 hierarchical categories. Extensive experiments on two real-world datasets demonstrate the superiority of our model over the state-of-the-art methods.

Due to the continuous growth of large-scale multi-modal data and increasing requirements for retrieval speed, deep cross-modal hashing has gained increasing attention recently. Most of existing studies take a similarity matrix as supervision to optimize their models, and the inner product between continuous surrogates of hash codes is utilized to depict the similarity in the Hamming space. However, all of them merely consider the relevant information to build the similarity matrix, ignoring the contribution of the irrelevant one, i.e., the categories that samples do not belong to. Therefore, they cannot effectively alleviate the effect of dissimilar samples. Moreover, due to the modality distribution difference, directly utilizing continuous surrogates of hash codes to calculate similarity may induce suboptimal retrieval performance. To tackle these issues, in this paper, we propose a novel deep normalized cross-modal hashing scheme with bi-direction relation reasoning, named Bi_NCMH. Specifically, we build the multi-level semantic similarity matrix by considering bi-direction relation, i.e., consistent and inconsistent relation. It hence can holistically characterize relations among instances. Besides, we execute feature normalization on continuous surrogates of hash codes to eliminate the deviation caused by modality gap, which further reduces the negative impact of binarization on retrieval performance. Extensive experiments on two cross-modal benchmark datasets demonstrate the superiority of our model over several state-of-the-art baselines.

This paper introduces SOAR: Spilling with Orthogonality-Amplified Residuals, a novel data indexing technique for approximate nearest neighbor (ANN) search. SOAR extends upon previous approaches to ANN search, such as spill trees, that utilize multiple redundant representations while partitioning the data to reduce the probability of missing a nearest neighbor during search. Rather than training and computing these redundant representations independently, however, SOAR uses an orthogonality-amplified residual loss, which optimizes each representation to compensate for cases where other representations perform poorly. This drastically improves the overall index quality, resulting in state-of-the-art ANN benchmark performance while maintaining fast indexing times and low memory consumption.

Cross-modal hashing encodes the multimedia data into a common binary hash space in which the correlations among the samples from different modalities can be effectively measured. Deep cross-modal hashing further improves the retrieval performance as the deep neural networks can generate more semantic relevant features and hash codes. In this paper, we study the unsupervised deep cross-modal hash coding and propose Deep Joint Semantics Reconstructing Hashing (DJSRH), which has the following two main advantages. First, to learn binary codes that preserve the neighborhood structure of the original data, DJSRH constructs a novel joint-semantics affinity matrix which elaborately integrates the original neighborhood information from different modalities and accordingly is capable to capture the latent intrinsic semantic affinity for the input multi-modal instances. Second, DJSRH later trains the networks to generate binary codes that maximally reconstruct above joint-semantics relations via the proposed reconstructing framework, which is more competent for the batch-wise training as it reconstructs the specific similarity value unlike the common Laplacian constraint merely preserving the similarity order. Extensive experiments demonstrate the significant improvement by DJSRH in various cross-modal retrieval tasks.

To convert the input into binary code, hashing algorithm has been widely used for approximate nearest neighbor search on large-scale image sets due to its computation and storage efficiency. Deep hashing further improves the retrieval quality by combining the hash coding with deep neural network. However, a major difficulty in deep hashing lies in the discrete constraints imposed on the network output, which generally makes the optimization NP hard. In this work, we adopt the greedy principle to tackle this NP hard problem by iteratively updating the network toward the probable optimal discrete solution in each iteration. A hash coding layer is designed to implement our approach which strictly uses the sign function in forward propagation to maintain the discrete constraints, while in back propagation the gradients are transmitted intactly to the front layer to avoid the vanishing gradients. In addition to the theoretical derivation, we provide a new perspective to visualize and understand the effectiveness and efficiency of our algorithm. Experiments on benchmark datasets show that our scheme outperforms state-of-the-art hashing methods in both supervised and unsupervised tasks.

Current state-of-the-art approximate nearest neighbor search (ANNS) algorithms generate indices that must be stored in main memory for fast high-recall search. This makes them expensive and limits the size of the dataset. We present a new graph-based indexing and search system called DiskANN that can index, store, and search a billion point database on a single workstation with just 64GB RAM and an inexpensive solid-state drive (SSD). Contrary to current wisdom, we demonstrate that the SSD-based indices built by DiskANN can meet all three desiderata for large-scale ANNS: high-recall, low query latency and high density (points indexed per node). On the billion point SIFT1B bigann dataset, DiskANN serves > 5000 queries a second with < 3ms mean latency and 95%+ 1-recall@1 on a 16 core machine, where state-of-the-art billion-point ANNS algorithms with similar memory footprint like FAISS and IVFOADC+G+P plateau at around 50% 1-recall@1. Alternately, in the high recall regime, DiskANN can index and serve 5 − 10x more points per node compared to state-of-the-art graph- based methods such as HNSW and NSG. Finally, as part of our overall DiskANN system, we introduce Vamana, a new graph-based ANNS index that is more versatile than the graph indices even for in-memory indices.

In recent years, binary coding techniques are becoming increasingly popular because of their high efficiency in handling large-scale computer vision applications. It has been demonstrated that supervised binary coding techniques that leverage supervised information can significantly enhance the coding quality, and hence greatly benefit visual search tasks. Typically, a modern binary coding method seeks to learn a group of coding functions which compress data samples into binary codes. However, few methods pursued the coding functions such that the precision at the top of a ranking list according to Hamming distances of the generated binary codes is optimized. In this paper, we propose a novel supervised binary coding approach, namely Top Rank Supervised Binary Coding (Top-RSBC), which explicitly focuses on optimizing the precision of top positions in a Hamming-distance ranking list towards preserving the supervision information. The core idea is to train the disciplined coding functions, by which the mistakes at the top of a Hamming-distance ranking list are penalized more than those at the bottom. To solve such coding functions, we relax the original discrete optimization objective with a continuous surrogate, and derive a stochastic gradient descent to optimize the surrogate objective. To further reduce the training time cost, we also design an online learning algorithm to optimize the surrogate objective more efficiently. Empirical studies based upon three benchmark image datasets demonstrate that the proposed binary coding approach achieves superior image search accuracy over the state-of-the-arts.

Existing video hash functions are built on three isolated stages: frame pooling, relaxed learning, and binarization, which have not adequately explored the temporal order of video frames in a joint binary optimization model, resulting in severe information loss. In this paper, we propose a novel unsupervised video hashing framework dubbed Self-Supervised Video Hashing (SSVH), that is able to capture the temporal nature of videos in an end-to-end learning-to-hash fashion. We specifically address two central problems: 1) how to design an encoder-decoder architecture to generate binary codes for videos; and 2) how to equip the binary codes with the ability of accurate video retrieval. We design a hierarchical binary autoencoder to model the temporal dependencies in videos with multiple granularities, and embed the videos into binary codes with less computations than the stacked architecture. Then, we encourage the binary codes to simultaneously reconstruct the visual content and neighborhood structure of the videos. Experiments on two real-world datasets (FCVID and YFCC) show that our SSVH method can significantly outperform the state-of-the-art methods and achieve the currently best performance on the task of unsupervised video retrieval.

In this paper, we present a new multimedia retrieval paradigm to innovate large-scale search of heterogenous multimedia data. It is able to return results of different media types from heterogeneous data sources, e.g., using a query image to retrieve relevant text documents or images from different data sources. This utilizes the widely available data from different sources and caters for the current users’ demand of receiving a result list simultaneously containing multiple types of data to obtain a comprehensive understanding of the query’s results. To enable large-scale inter-media retrieval, we propose a novel inter-media hashing (IMH) model to explore the correlations among multiple media types from different data sources and tackle the scalability issue. To this end, multimedia data from heterogeneous data sources are transformed into a common Hamming space, in which fast search can be easily implemented by XOR and bit-count operations. Furthermore, we integrate a linear regression model to learn hashing functions so that the hash codes for new data points can be efficiently generated. Experiments conducted on real-world large-scale multimedia datasets demonstrate the superiority of our proposed method compared with state-of-the-art techniques.

Customizable image retrieval from large datasets remains a critical challenge, particularly when preserving spatial relationships within images. Traditional hashing methods, primarily based on deep learning, often fail to capture spatial information adequately and lack transparency. In this paper, we introduce NeuroHash, a novel neuro-symbolic framework leveraging Hyperdimensional Computing (HDC) to enable highly customizable, spatially-aware image retrieval. NeuroHash combines pre-trained deep neural network models with HDC-based symbolic models, allowing for flexible manipulation of hash values to support conditional image retrieval. Our method includes a self-supervised context-aware HDC encoder and novel loss terms for optimizing lower-dimensional bipolar hashing using multilinear hyperplanes. We evaluate NeuroHash on two benchmark datasets, demonstrating superior performance compared to state-of-the-art hashing methods, as measured by mAP@5K scores and our newly introduced metric, mAP@5Kr, which assesses spatial alignment. The results highlight NeuroHash’s ability to achieve competitive performance while offering significant advantages in flexibility and customization, paving the way for more advanced and versatile image retrieval systems.

Due to its low storage cost and fast query speed, hashing has been widely adopted for approximate nearest neighbor search in large-scale datasets. Traditional hashing methods try to learn the hash codes in an unsupervised way where the metric (Euclidean) structure of the training data is preserved. Very recently, supervised hashing methods, which try to preserve the semantic structure constructed from the semantic labels of the training points, have exhibited higher accuracy than unsupervised methods. In this paper, we propose a novel supervised hashing method, called latent factor hashing (LFH), to learn similarity-preserving binary codes based on latent factor models. An algorithm with convergence guarantee is proposed to learn the parameters of LFH. Furthermore, a linear-time variant with stochastic learning is proposed for training LFH on large-scale datasets. Experimental results on two large datasets with semantic labels show that LFH can achieve superior accuracy than state-of-the-art methods with comparable training time.

The ability of fast similarity search at large scale is of great importance to many Information Retrieval (IR) applications. A promising way to accelerate similarity search is semantic hashing which designs compact binary codes for a large number of documents so that semantically similar documents are mapped to similar codes (within a short Hamming distance). Although some recently proposed techniques are able to generate high-quality codes for documents known in advance, obtaining the codes for previously unseen documents remains to be a very challenging problem. In this paper, we emphasise this issue and propose a novel SelfTaught Hashing (STH) approach to semantic hashing: we first find the optimal l-bit binary codes for all documents in the given corpus via unsupervised learning, and then train l classifiers via supervised learning to predict the l-bit code for any query document unseen before. Our experiments on three real-world text datasets show that the proposed approach using binarised Laplacian Eigenmap (LapEig) and linear Support Vector Machine (SVM) outperforms stateof-the-art techniques significantly.

Fast image retrieval is of great importance in many computer vision tasks and especially practical applications. Deep hashing, the state-of-the-art fast image retrieval scheme, introduces deep learning to learn the hash functions and generate binary hash codes, and outperforms the other image retrieval methods in terms of accuracy. However, all the existing deep hashing methods could only generate one level hash codes and require a linear traversal of all the hash codes to figure out the closest one when a new query arrives, which is very time-consuming and even intractable for large scale applications. In this work, we propose a Hierarchical Deep Hashing(HDHash) scheme to speed up the state-of-the-art deep hashing methods. More specifically, hierarchical deep hash codes of multiple levels can be generated and indexed with tree structures rather than linear ones, and pruning irrelevant branches can sharply decrease the retrieval time. To our best knowledge, this is the first work to introduce hierarchical indexed deep hashing for fast large scale image retrieval. Extensive experimental results on three benchmark datasets demonstrate that the proposed HDHash scheme achieves better or comparable accuracy with significantly improved efficiency and reduced memory as compared to state-of- the-art fast image retrieval schemes.

In recent years, cross-media retrieval has drawn considerable attention due to the exponential growth of multimedia data. Many hashing approaches have been proposed for the cross-media search task. However, there are still open problems that warrant investigation. For example, most existing supervised hashing approaches employ a binary label matrix, which achieves small margins between wrong labels (0) and true labels (1). This may affect the retrieval performance by generating many false negatives and false positives. In addition, some methods adopt a relaxation scheme to solve the binary constraints, which may cause large quantization errors. There are also some discrete hashing methods that have been presented, but most of them are time-consuming. To conquer these problems, we present a label relaxation and discrete matrix factorization method (LRMF) for cross-modal retrieval. It offers a number of innovations. First of all, the proposed approach employs a novel label relaxation scheme to control the margins adaptively, which has the benefit of reducing the quantization error. Second, by virtue of the proposed discrete matrix factorization method designed to learn the binary codes, large quantization errors caused by relaxation can be avoided. The experimental results obtained on two widely-used databases demonstrate that LRMF outperforms state-of-the-art cross-media methods.

In industrial recommendation systems on websites and apps, it is essential to recall and predict top-n results relevant to user interests from a content pool of billions within milliseconds. To cope with continuous data growth and improve real-time recommendation performance, we have designed and implemented a high-performance batch query architecture for real-time recommendation systems. Our contributions include optimizing hash structures with a cacheline-aware probing method to enhance coalesced hashing, as well as the implementation of a hybrid storage key-value service built upon it. Our experiments indicate this approach significantly surpasses conventional hash tables in batch query throughput, achieving up to 90% of the query throughput of random memory access when incorporating parallel optimization. The support for NVMe, integrating two-tier storage for hot and cold data, notably reduces resource consumption. Additionally, the system facilitates dynamic updates, automated sharding of attributes and feature embedding tables, and introduces innovative protocols for consistency in batch queries, thereby enhancing the effectiveness of real-time incremental learning updates. This architecture has been deployed and in use in the bilibili recommendation system for over a year, a video content community with hundreds of millions of users, supporting 10x increase in model computation with minimal resource growth, improving outcomes while preserving the system’s real-time performance.

Implicit neural representations (INRs) recently achieved great success in image representation and compression, offering high visual quality and fast rendering speeds with 10-1000 FPS, assuming sufficient GPU resources are available. However, this requirement often hinders their use on low-end devices with limited memory. In response, we propose a groundbreaking paradigm of image representation and compression by 2D Gaussian Splatting, named GaussianImage. We first introduce 2D Gaussian to represent the image, where each Gaussian has 8 parameters including position, covariance and color. Subsequently, we unveil a novel rendering algorithm based on accumulated summation. Remarkably, our method with a minimum of 3\(\times\) lower GPU memory usage and 5\(\times\) faster fitting time not only rivals INRs (e.g., WIRE, I-NGP) in representation performance, but also delivers a faster rendering speed of 1500-2000 FPS regardless of parameter size. Furthermore, we integrate existing vector quantization technique to build an image codec. Experimental results demonstrate that our codec attains rate-distortion performance comparable to compression-based INRs such as COIN and COIN++, while facilitating decoding speeds of approximately 2000 FPS. Additionally, preliminary proof of concept shows that our codec surpasses COIN and COIN++ in performance when using partial bits-back coding. Code is available at https://github.com/Xinjie-Q/GaussianImage.

In light of the ability to enable efficient storage and fast query for big data, hashing techniques for cross-modal search have aroused extensive attention. Despite the great success achieved, unsupervised cross-modal hashing still suffers from lacking reliable similarity supervision and struggles with handling the heterogeneity issue between different modalities. To cope with these, in this paper, we devise a new deep hashing model, termed as High-order Nonlocal Hashing (HNH) to facilitate cross-modal retrieval with the following advantages. First, different from existing methods that mainly leverage low-level local-view similarity as the guidance for hashing learning, we propose a high-order affinity measure that considers the multi-modal neighbourhood structures from a nonlocal perspective, thereby comprehensively capturing the similarity relationships between data items. Second, a common representation is introduced to correlate different modalities. By enforcing the modal-specific descriptors and the common representation to be aligned with each other, the proposed HNH significantly bridges the modality gap and maintains the intra-consistency. Third, an effective affinity preserving objective function is delicately designed to generate high-quality binary codes. Extensive experiments evidence the superiority of the proposed HNH in unsupervised cross-modal retrieval tasks over the state-of-the-art baselines.

With the advantages of low storage cost and extremely fast retrieval speed, deep hashing methods have attracted much attention for image retrieval recently. However, large-scale face image retrieval with significant intra-class variations is still challenging. Neither existing pairwise/triplet labels-based nor softmax classification loss-based deep hashing works can generate compact and discriminative binary codes. Considering these issues, we propose a center-based framework integrating end-to-end hashing learning and class centers learning simultaneously. The framework minimizes the intra-class variance by clustering intra-class samples into a learnable class center. To strengthen inter-class separability, it additionally imposes a novel regularization term to enlarge the Hamming distance between pairwise class centers. Moreover, a simple yet effective regression matrix is introduced to encourage intra-class samples to generate the same binary codes, which further enhances the hashing codes compactness. Experiments on four large-scale datasets show the proposed method outperforms state-of-the-art baselines under various code lengths and commonly-used evaluation metrics.

Similarity search applications with a large amount of text and image data demands an efficient and effective solution. One useful strategy is to represent the examples in databases as compact binary codes through semantic hashing, which has attracted much attention due to its fast query/search speed and drastically reduced storage requirement. All of the current semantic hashing methods only deal with the case when each example is represented by one type of features. However, examples are often described from several different information sources in many real world applications. For example, the characteristics of a webpage can be derived from both its content part and its associated links. To address the problem of learning good hashing codes in this scenario, we propose a novel research problem – Composite Hashing with Multiple Information Sources (CHMIS). The focus of the new research problem is to design an algorithm for incorporating the features from different information sources into the binary hashing codes efficiently and effectively. In particular, we propose an algorithm CHMISAW (CHMIS with Adjusted Weights) for learning the codes. The proposed algorithm integrates information from several different sources into the binary hashing codes by adjusting the weights on each individual source for maximizing the coding performance, and enables fast conversion from query examples to their binary hashing codes. Experimental results on five different datasets demonstrate the superior performance of the proposed method against several other state-of-the-art semantic hashing techniques.

Image-text matching has been a long-standing problem, which seeks to connect vision and language through semantic understanding. Due to the capability to manage large-scale raw data, unsupervised hashing-based approaches have gained prominence recently. They typically construct a semantic similarity structure using the natural distance, which subsequently provides guidance to the model optimization process. However, the similarity structure could be biased at the boundaries of semantic distributions, causing error accumulation during sequential optimization. To tackle this, we introduce a novel hashing approach termed Distribution-based Structure Mining with Consistency Learning (DEMO) for efficient image-text matching. From a statistical view, DEMO characterizes each image using multiple augmented views, which are considered as samples drawn from its intrinsic semantic distribution. Then, we employ a non-parametric distribution divergence to ensure a robust and precise similarity structure. In addition, we introduce collaborative consistency learning which not only preserves the similarity structure in the Hamming space but also encourages consistency between retrieval distribution from different directions in a self-supervised manner. Through extensive experiments on three benchmark image-text matching datasets, we demonstrate that DEMO achieves superior performance compared with many state-of-the-art methods.

Binary hashing has been widely used for efficient similarity search due to its query and storage efficiency. In most existing binary hashing methods, the high-dimensional data are embedded into Hamming space and the distance or similarity of two points are approximated by the Hamming distance between their binary codes. The Hamming distance calculation is efficient, however, in practice, there are often lots of results sharing the same Hamming distance to a query, which makes this distance measure ambiguous and poses a critical issue for similarity search where ranking is important. In this paper, we propose a weighted Hamming distance ranking algorithm (WhRank) to rank the binary codes of hashing methods. By assigning different bit-level weights to different hash bits, the returned binary codes are ranked at a finer-grained binary code level. We give an algorithm to learn the data-adaptive and query-sensitive weight for each hash bit. Evaluations on two large-scale image data sets demonstrate the efficacy of our weighted Hamming distance for binary code ranking.

Extracting informative image features and learning effective approximate hashing functions are two crucial steps in image retrieval . Conventional methods often study these two steps separately, e.g., learning hash functions from a predefined hand-crafted feature space. Meanwhile, the bit lengths of output hashing codes are preset in most previous methods, neglecting the significance level of different bits and restricting their practical flexibility. To address these issues, we propose a supervised learning framework to generate compact and bit-scalable hashing codes directly from raw images. We pose hashing learning as a problem of regularized similarity learning. Specifically, we organize the training images into a batch of triplet samples, each sample containing two images with the same label and one with a different label. With these triplet samples, we maximize the margin between matched pairs and mismatched pairs in the Hamming space. In addition, a regularization term is introduced to enforce the adjacency consistency, i.e., images of similar appearances should have similar codes. The deep convolutional neural network is utilized to train the model in an end-to-end fashion, where discriminative image features and hash functions are simultaneously optimized. Furthermore, each bit of our hashing codes is unequally weighted so that we can manipulate the code lengths by truncating the insignificant bits. Our framework outperforms state-of-the-arts on public benchmarks of similar image search and also achieves promising results in the application of person re-identification in surveillance. It is also shown that the generated bit-scalable hashing codes well preserve the discriminative powers with shorter code lengths.

In this paper, we propose training very deep neural networks (DNNs) for supervised learning of hash codes. Existing methods in this context train relatively “shallow” networks limited by the issues arising in back propagation (e.e. vanishing gradients) as well as computational efficiency. We propose a novel and efficient training algorithm inspired by alternating direction method of multipliers (ADMM) that overcomes some of these limitations. Our method decomposes the training process into independent layer-wise local updates through auxiliary variables. Empirically we observe that our training algorithm always converges and its computational complexity is linearly proportional to the number of edges in the networks. Empirically we manage to train DNNs with 64 hidden layers and 1024 nodes per layer for supervised hashing in about 3 hours using a single GPU. Our proposed very deep supervised hashing (VDSH) method significantly outperforms the state-of-the-art on several benchmark datasets.

Due to its low storage cost and fast query speed, hashing has been widely adopted for similarity search in multimedia data. In particular, more and more attentions have been payed to multimodal hashing for search in multimedia data with multiple modalities, such as images with tags. Typically, supervised information of semantic labels is also available for the data points in many real applications. Hence, many supervised multimodal hashing (SMH) methods have been proposed to utilize such semantic labels to further improve the search accuracy. However, the training time complexity of most existing SMH methods is too high, which makes them unscalable to large-scale datasets. In this paper, a novel SMH method, called semantic correlation maximization (SCM), is proposed to seamlessly integrate semantic labels into the hashing learning procedure for large-scale data modeling. Experimental results on two real-world datasets show that SCM can signifi- cantly outperform the state-of-the-art SMH methods, in terms of both accuracy and scalability.

With the rapid growth of web images, hashing has received increasing interests in large scale image retrieval. Research efforts have been devoted to learning compact binary codes that preserve semantic similarity based on labels. However, most of these hashing methods are designed to handle simple binary similarity. The complex multilevel semantic structure of images associated with multiple labels have not yet been well explored. Here we propose a deep semantic ranking based method for learning hash functions that preserve multilevel semantic similarity between multilabel images. In our approach, deep convolutional neural network is incorporated into hash functions to jointly learn feature representations and mappings from them to hash codes, which avoids the limitation of semantic representation power of hand-crafted features. Meanwhile, a ranking list that encodes the multilevel similarity information is employed to guide the learning of such deep hash functions. An effective scheme based on surrogate loss is used to solve the intractable optimization problem of nonsmooth and multivariate ranking measures involved in the learning procedure. Experimental results show the superiority of our proposed approach over several state-of-theart hashing methods in term of ranking evaluation metrics when tested on multi-label image datasets.

We present a probabilistic framework for learning pairwise similarities between objects belonging to different modalities, such as drugs and proteins, or text and images. Our framework is based on learning a binary code based representation for objects in each modality, and has the following key properties: (i) it can leverage both pairwise as well as easy-to-obtain relative preference based cross-modal constraints, (ii) the probabilistic framework naturally allows querying for the most useful/informative constraints, facilitating an active learning setting (existing methods for cross-modal similarity learning do not have such a mechanism), and (iii) the binary code length is learned from the data. We demonstrate the effectiveness of the proposed approach on two problems that require computing pairwise similarities between cross-modal object pairs: cross-modal link prediction in bipartite graphs, and hashing based cross-modal similarity search.

Hashing-based methods provide a very promising approach to large-scale similarity search. To obtain compact hash codes, a recent trend seeks to learn the hash functions from data automatically. In this paper, we study hash function learning in the context of multimodal data. We propose a novel multimodal hash function learning method, called Co-Regularized Hashing (CRH), based on a boosted coregularization framework. The hash functions for each bit of the hash codes are learned by solving DC (difference of convex functions) programs, while the learning for multiple bits proceeds via a boosting procedure so that the bias introduced by the hash functions can be sequentially minimized. We empirically compare CRH with two state-of-the-art multimodal hash function learning methods on two publicly available data sets.

Due to the storage and retrieval efficiency, hashing has been widely deployed to approximate nearest neighbor search for large-scale multimedia retrieval. Supervised hashing, which improves the quality of hash coding by exploiting the semantic similarity on data pairs, has received increasing attention recently. For most existing supervised hashing methods for image retrieval, an image is first represented as a vector of hand-crafted or machine-learned features, followed by another separate quantization step that generates binary codes. However, suboptimal hash coding may be produced, because the quantization error is not statistically minimized and the feature representation is not optimally compatible with the binary coding. In this paper, we propose a novel Deep Hashing Network (DHN) architecture for supervised hashing, in which we jointly learn good image representation tailored to hash coding and formally control the quantization error. The DHN model constitutes four key components: (1) a sub-network with multiple convolution-pooling layers to capture image representations; (2) a fully-connected hashing layer to generate compact binary hash codes; (3) a pairwise cross-entropy loss layer for similarity-preserving learning; and (4) a pairwise quantization loss for controlling hashing quality. Extensive experiments on standard image retrieval datasets show the proposed DHN model yields substantial boosts over latest state-of-the-art hashing methods.

Most existing cross-modal hashing methods suffer from the scalability issue in the training phase. In this paper, we propose a novel cross-modal hashing approach with a linear time complexity to the training data size, to enable scalable indexing for multimedia search across multiple modals. Taking both the intra-similarity in each modal and the inter-similarity across different modals into consideration, the proposed approach aims at effectively learning hash functions from large-scale training datasets. More specifically, for each modal, we first partition the training data into \(k\) clusters and then represent each training data point with its distances to \(k\) centroids of the clusters. Interestingly, such a k-dimensional data representation can reduce the time complexity of the training phase from traditional O(n2) or higher to O(n), where \(n\) is the training data size, leading to practical learning on large-scale datasets. We further prove that this new representation preserves the intra-similarity in each modal. To preserve the inter-similarity among data points across different modals, we transform the derived data representations into a common binary subspace in which binary codes from all the modals are “consistent” and comparable. The transformation simultaneously outputs the hash functions for all modals, which are used to convert unseen data into binary codes. Given a query of one modal, it is first mapped into the binary codes using the modal’s hash functions, followed by matching the database binary codes of any other modals. Experimental results on two benchmark datasets confirm the scalability and the effectiveness of the proposed approach in comparison with the state of the art.

Automatic crash bucketing is a crucial phase in the software development process for efficiently triaging bug reports. It generally consists in grouping similar reports through clustering techniques. However, with real-time streaming bug collection, systems are needed to quickly answer the question: What are the most similar bugs to a new one?, that is, efficiently find near-duplicates. It is thus natural to consider nearest neighbors search to tackle this problem and especially the well-known locality-sensitive hashing (LSH) to deal with large datasets due to its sublinear performance and theoretical guarantees on the similarity search accuracy. Surprisingly, LSH has not been considered in the crash bucketing literature. It is indeed not trivial to derive hash functions that satisfy the so-called locality-sensitive property for the most advanced crash bucketing metrics. Consequently, we study in this paper how to leverage LSH for this task. To be able to consider the most relevant metrics used in the literature, we introduce DeepLSH, a Siamese DNN architecture with an original loss function, that perfectly approximates the locality-sensitivity property even for Jaccard and Cosine metrics for which exact LSH solutions exist. We support this claim with a series of experiments on an original dataset, which we make available.

Identifying texts with a given semantics is central for many information seeking scenarios. Similarity search over vector embeddings appear to be central to this ability, yet the similarity reflected in current text embeddings is corpus-driven, and is inconsistent and sub-optimal for many use cases. What, then, is a good notion of similarity for effective retrieval of text? We identify the need to search for texts based on abstract descriptions of their content, and the corresponding notion of description based similarity. We demonstrate the inadequacy of current text embeddings and propose an alternative model that significantly improves when used in standard nearest neighbor search. The model is trained using positive and negative pairs sourced through prompting a LLM, demonstrating how data from LLMs can be used for creating new capabilities not immediately possible using the original model.

Our research investigates the recommendation of code examples to aid software developers, a practice that saves developers significant time by providing ready-to-use code snippets. The focus of our study is Stack Overflow, a commonly used resource for coding discussions and solutions, particularly in the context of the Java programming language. We applied BERT, a powerful Large Language Model (LLM) that enables us to transform code examples into numerical vectors by extracting their semantic information. Once these numerical representations are prepared, we identify Approximate Nearest Neighbors (ANN) using Locality-Sensitive Hashing (LSH). Our research employed two variants of LSH: Random Hyperplane-based LSH and Query-Aware LSH. We rigorously compared these two approaches across four parameters: HitRate, Mean Reciprocal Rank (MRR), Average Execution Time, and Relevance. Our study revealed that the Query-Aware (QA) approach showed superior performance over the Random Hyperplane-based (RH) method. Specifically, it exhibited a notable improvement of 20\% to 35\% in HitRate for query pairs compared to the RH approach. Furthermore, the QA approach proved significantly more time-efficient, with its speed in creating hashing tables and assigning data samples to buckets being at least four times faster. It can return code examples within milliseconds, whereas the RH approach typically requires several seconds to recommend code examples. Due to the superior performance of the QA approach, we tested it against PostFinder and FaCoY, the state-of-the-art baselines. Our QA method showed comparable efficiency proving its potential for effective code recommendation.

Locality-sensitive hashing (LSH) based frameworks have been used efficiently to select weight vectors in a dense hidden layer with high cosine similarity to an input, enabling dynamic pruning. While this type of scheme has been shown to improve computational training efficiency, existing algorithms require repeated randomized projection of the full layer weight, which is impractical for computational- and memory-constrained devices. In a distributed setting, deferring LSH analysis to a centralized host is (i) slow if the device cluster is large and (ii) requires access to input data which is forbidden in a federated context. Using a new family of hash functions, we develop one of the first private, personalized, and memory-efficient on-device LSH frameworks. Our framework enables privacy and personalization by allowing each device to generate hash tables, without the help of a central host, using device-specific hashing hyper-parameters (e.g. number of hash tables or hash length). Hash tables are generated with a compressed set of the full weights, and can be serially generated and discarded if the process is memory-intensive. This allows devices to avoid maintaining (i) the fully-sized model and (ii) large amounts of hash tables in local memory for LSH analysis. We prove several statistical and sensitivity properties of our hash functions, and experimentally demonstrate that our framework is competitive in training large-scale recommender networks compared to other LSH frameworks which assume unrestricted on-device capacity.

Open-vocabulary querying in 3D space is challenging but essential for scene understanding tasks such as object localization and segmentation. Language-embedded scene representations have made progress by incorporating language features into 3D spaces. However, their efficacy heavily depends on neural networks that are resource-intensive in training and rendering. Although recent 3D Gaussians offer efficient and high-quality novel view synthesis, directly embedding language features in them leads to prohibitive memory usage and decreased performance. In this work, we introduce Language Embedded 3D Gaussians, a novel scene representation for open-vocabulary query tasks. Instead of embedding high-dimensional raw semantic features on 3D Gaussians, we propose a dedicated quantization scheme that drastically alleviates the memory requirement, and a novel embedding procedure that achieves smoother yet high accuracy query, countering the multi-view feature inconsistencies and the high-frequency inductive bias in point-based representations. Our comprehensive experiments show that our representation achieves the best visual quality and language querying accuracy across current language-embedded representations, while maintaining real-time rendering frame rates on a single desktop GPU.

Hashing is at the heart of large-scale image similarity search, and recent methods have been substantially improved through deep learning techniques. Such algorithms typically learn continuous embeddings of the data. To avoid a subsequent costly binarization step, a common solution is to employ loss functions that combine a similarity learning term (to ensure similar images are grouped to nearby embeddings) and a quantization penalty term (to ensure that the embedding entries are close to binarized entries, e.g., -1 or 1). Still, the interaction between these two terms can make learning harder and the embeddings worse. We propose an alternative quantization strategy that decomposes the learning problem in two stages: first, perform similarity learning over the embedding space with no quantization; second, find an optimal orthogonal transformation of the embeddings so each coordinate of the embedding is close to its sign, and then quantize the transformed embedding through the sign function. In the second step, we parametrize orthogonal transformations using Householder matrices to efficiently leverage stochastic gradient descent. Since similarity measures are usually invariant under orthogonal transformations, this quantization strategy comes at no cost in terms of performance. The resulting algorithm is unsupervised, fast, hyperparameter-free and can be run on top of any existing deep hashing or metric learning algorithm. We provide extensive experimental results showing that this approach leads to state-of-the-art performance on widely used image datasets, and, unlike other quantization strategies, brings consistent improvements in performance to existing deep hashing algorithms.

Two-tower models are a prevalent matching framework for recommendation, which have been widely deployed in industrial applications. The success of two-tower matching attributes to its efficiency in retrieval among a large number of items, since the item tower can be precomputed and used for fast Approximate Nearest Neighbor (ANN) search. However, it suffers two main challenges, including limited feature interaction capability and reduced accuracy in online serving. Existing approaches attempt to design novel late interactions instead of dot products, but they still fail to support complex feature interactions or lose retrieval efficiency. To address these challenges, we propose a new matching paradigm named SparCode, which supports not only sophisticated feature interactions but also efficient retrieval. Specifically, SparCode introduces an all-to-all interaction module to model fine-grained query-item interactions. Besides, we design a discrete code-based sparse inverted index jointly trained with the model to achieve effective and efficient model inference. Extensive experiments have been conducted on open benchmark datasets to demonstrate the superiority of our framework. The results show that SparCode significantly improves the accuracy of candidate item matching while retaining the same level of retrieval efficiency with two-tower models. Our source code will be available at MindSpore/models.

Dense retrieval models use bi-encoder network architectures for learning query and document representations. These representations are often in the form of a vector representation and their similarities are often computed using the dot product function. In this paper, we propose a new representation learning framework for dense retrieval. Instead of learning a vector for each query and document, our framework learns a multivariate distribution and uses negative multivariate KL divergence to compute the similarity between distributions. For simplicity and efficiency reasons, we assume that the distributions are multivariate normals and then train large language models to produce mean and variance vectors for these distributions. We provide a theoretical foundation for the proposed framework and show that it can be seamlessly integrated into the existing approximate nearest neighbor algorithms to perform retrieval efficiently. We conduct an extensive suite of experiments on a wide range of datasets, and demonstrate significant improvements compared to competitive dense retrieval models.

Randomly-hashed item ids are used ubiquitously in recommendation models. However, the learned representations from random hashing prevents generalization across similar items, causing problems of learning unseen and long-tail items, especially when item corpus is large, power-law distributed, and evolving dynamically. In this paper, we propose using content-derived features as a replacement for random ids. We show that simply replacing ID features with content-based embeddings can cause a drop in quality due to reduced memorization capability. To strike a good balance of memorization and generalization, we propose to use Semantic IDs – a compact discrete item representation learned from frozen content embeddings using RQ-VAE that captures the hierarchy of concepts in items – as a replacement for random item ids. Similar to content embeddings, the compactness of Semantic IDs poses a problem of easy adaption in recommendation models. We propose novel methods for adapting Semantic IDs in industry-scale ranking models, through hashing sub-pieces of of the Semantic-ID sequences. In particular, we find that the SentencePiece model that is commonly used in LLM tokenization outperforms manually crafted pieces such as N-grams. To the end, we evaluate our approaches in a real-world ranking model for YouTube recommendations. Our experiments demonstrate that Semantic IDs can replace the direct use of video IDs by improving the generalization ability on new and long-tail item slices without sacrificing overall model quality.

Visual Place recognition is commonly addressed as an image retrieval problem. However, retrieval methods are impractical to scale to large datasets, densely sampled from city-wide maps, since their dimension impact negatively on the inference time. Using approximate nearest neighbour search for retrieval helps to mitigate this issue, at the cost of a performance drop. In this paper we investigate whether we can effectively approach this task as a classification problem, thus bypassing the need for a similarity search. We find that existing classification methods for coarse, planet-wide localization are not suitable for the fine-grained and city-wide setting. This is largely due to how the dataset is split into classes, because these methods are designed to handle a sparse distribution of photos and as such do not consider the visual aliasing problem across neighbouring classes that naturally arises in dense scenarios. Thus, we propose a partitioning scheme that enables a fast and accurate inference, preserving a simple learning procedure, and a novel inference pipeline based on an ensemble of novel classifiers that uses the prototypes learned via an angular margin loss. Our method, Divide&Classify (D&C), enjoys the fast inference of classification solutions and an accuracy competitive with retrieval methods on the fine-grained, city-wide setting. Moreover, we show that D&C can be paired with existing retrieval pipelines to speed up computations by over 20 times while increasing their recall, leading to new state-of-the-art results.

In the first part of this survey, we review how the power of two choices underlies space-efficient data structures like cuckoo hash tables. We’ll find that the additional power afforded by more than 2 choices is often outweighed by the additional costs they bring. In the second part, we present a data structure where choices play a role at coarser than per-element granularity. In some sense, we rely on the power of \(1+\epsilon\) choices.

Most of the research in content-based image retrieval (CBIR) focus on developing robust feature representations that can effectively retrieve instances from a database of images that are visually similar to a query. However, the retrieved images sometimes contain results that are not semantically related to the query. To address this, we propose a method for CBIR that captures both visual and semantic similarity using a visual hierarchy. The hierarchy is constructed by merging classes with overlapping features in the latent space of a deep neural network trained for classification, assuming that overlapping classes share high visual and semantic similarities. Finally, the constructed hierarchy is integrated into the distance calculation metric for similarity search. Experiments on standard datasets: CUB-200-2011 and CIFAR100, and a real-life use case using diatom microscopy images show that our method achieves superior performance compared to the existing methods on image retrieval.

These are the lecture notes for the course CM0622 - Algorithms for Massive Data, Ca’ Foscari University of Venice. The goal of this course is to introduce algorithmic techniques for dealing with massive data: data so large that it does not fit in the computer’s memory. There are two main solutions to deal with massive data: (lossless) compressed data structures and (lossy) data sketches. These notes cover both topics: compressed suffix arrays, probabilistic filters, sketching under various metrics, Locality Sensitive Hashing, nearest neighbour search, algorithms on streams (pattern matching, counting).

The ongoing Big Data explosion has created a demand for efficient and scalable algorithms for similarity search. Most recent work has focused on \textit{approximate} \(k\)-NN search, and while this may be sufficient for some applications, \textit{exact} \(k\)-NN search would be ideal for many applications. We present CAKES, a set of three novel, exact algorithms for \(k\)-NN search. CAKES’s algorithms are generic over \textit{any} distance function, and they \textit{do not} scale with the cardinality or embedding dimension of the dataset, but rather with its metric entropy and fractal dimension. We test these claims on datasets from the ANN-Benchmarks suite under commonly-used distance functions, as well as on a genomic dataset with Levenshtein distance and a radio-frequency dataset with Dynamic Time Warping distance. We demonstrate that CAKES exhibits near-constant scaling with cardinality on data conforming to the manifold hypothesis, and has perfect recall on data in \textit{metric} spaces. We also demonstrate that CAKES exhibits significantly higher recall than state-of-the-art \(k\)-NN search algorithms when the distance function is not a metric. Additionally, we show that indexing and tuning time for CAKES is an order of magnitude, or more, faster than state-of-the-art approaches. We conclude that CAKES is a highly efficient and scalable algorithm for exact \(k\)-NN search on Big Data. We provide a Rust implementation of CAKES.

Online string matching is a computational problem involving the search for patterns or substrings in a large text dataset, with the pattern and text being processed sequentially, without prior access to the entire text. Its relevance stems from applications in data compression, data mining, text editing, and bioinformatics, where rapid and efficient pattern matching is crucial. Various solutions have been proposed over the past few decades, employing diverse techniques. Recently, weak recognition approaches have attracted increasing attention. This paper presents Hash Chain, a new algorithm based on a robust weak factor recognition approach that connects adjacent factors through hashing. Despite its O(nm) complexity, the algorithm exhibits a sublinear behavior in practice and achieves superior performance compared to the most effective algorithms.

Approximate Nearest Neighbor Search (ANNS) plays a critical role in various disciplines spanning data mining and artificial intelligence, from information retrieval and computer vision to natural language processing and recommender systems. Data volumes have soared in recent years and the computational cost of an exhaustive exact nearest neighbor search is often prohibitive, necessitating the adoption of approximate techniques. The balanced performance and recall of graph-based approaches have more recently garnered significant attention in ANNS algorithms, however, only a few studies have explored harnessing the power of GPUs and multi-core processors despite the widespread use of massively parallel and general-purpose computing. To bridge this gap, we introduce a novel parallel computing hardware-based proximity graph and search algorithm. By leveraging the high-performance capabilities of modern hardware, our approach achieves remarkable efficiency gains. In particular, our method surpasses existing CPU and GPU-based methods in constructing the proximity graph, demonstrating higher throughput in both large- and small-batch searches while maintaining compatible accuracy. In graph construction time, our method, CAGRA, is 2.2~27x faster than HNSW, which is one of the CPU SOTA implementations. In large-batch query throughput in the 90% to 95% recall range, our method is 33~77x faster than HNSW, and is 3.8~8.8x faster than the SOTA implementations for GPU. For a single query, our method is 3.4~53x faster than HNSW at 95% recall.

Cryptocurrency mining processes always lead to a high energy consumption at considerably high production cost, which is nearly one-third of cryptocurrency (e.g. Bitcoin) price itself. As the core of mining process is based on SHA-256 cryptographic hashing function, by using the alternative quantum computers, hybrid quantum computers or more larger quantum computing devices like quantum annealers, it would be possible to reduce the mining energy consumption with a quantum hardware’s low-energy-operation characteristics. Within this work we demonstrated the use of optimized quantum mining facilities which would replace the classical SHA-256 and high energy consuming classical hardware in near future.

Biometric applications, such as person re-identification (ReID), are often deployed on energy constrained devices. While recent ReID methods prioritize high retrieval performance, they often come with large computational costs and high search time, rendering them less practical in real-world settings. In this work, we propose an input-adaptive network with multiple exit blocks, that can terminate computation early if the retrieval is straightforward or noisy, saving a lot of computation. To assess the complexity of the input, we introduce a temporal-based classifier driven by a new training strategy. Furthermore, we adopt a binary hash code generation approach instead of relying on continuous-valued features, which significantly improves the search process by a factor of 20. To ensure similarity preservation, we utilize a new ranking regularizer that bridges the gap between continuous and binary features. Extensive analysis of our proposed method is conducted on three datasets: Market1501, MSMT17 (Multi-Scene Multi-Time), and the BGC1 (BRIAR Government Collection). Using our approach, more than 70% of the samples with compact hash codes exit early on the Market1501 dataset, saving 80% of the networks computational cost and improving over other hash-based methods by 60%. These results demonstrate a significant improvement over dynamic networks and showcase comparable accuracy performance to conventional ReID methods. Code will be made available.

Approximate Nearest Neighbor Search (ANNS) is the task of finding the database vector that is closest to a given query vector. Graph-based ANNS is the family of methods with the best balance of accuracy and speed for million-scale datasets. However, graph-based methods have the disadvantage of long index construction time. Recently, many researchers have improved the tradeoff between accuracy and speed during a search. However, there is little research on accelerating index construction. We propose a fast graph construction algorithm, Relative NN-Descent (RNN-Descent). RNN-Descent combines NN-Descent, an algorithm for constructing approximate K-nearest neighbor graphs (K-NN graphs), and RNG Strategy, an algorithm for selecting edges effective for search. This algorithm allows the direct construction of graph-based indexes without ANNS. Experimental results demonstrated that the proposed method had the fastest index construction speed, while its search performance is comparable to existing state-of-the-art methods such as NSG. For example, in experiments on the GIST1M dataset, the construction of the proposed method is 2x faster than NSG. Additionally, it was even faster than the construction speed of NN-Descent.

Unsupervised hashing methods typically aim to preserve the similarity between data points in a feature space by mapping them to binary hash codes. However, these methods often overlook the fact that the similarity between data points in the continuous feature space may not be preserved in the discrete hash code space, due to the limited similarity range of hash codes. The similarity range is bounded by the code length and can lead to a problem known as similarity collapse. That is, the positive and negative pairs of data points become less distinguishable from each other in the hash space. To alleviate this problem, in this paper a novel Similarity Distribution Calibration (SDC) method is introduced. SDC aligns the hash code similarity distribution towards a calibration distribution (e.g., beta distribution) with sufficient spread across the entire similarity range, thus alleviating the similarity collapse problem. Extensive experiments show that our SDC outperforms significantly the state-of-the-art alternatives on coarse category-level and instance-level image retrieval. Code is available at https://github.com/kamwoh/sdc.

Retrieving relevant items that match users’ queries from billion-scale corpus forms the core of industrial e-commerce search systems, in which embedding-based retrieval (EBR) methods are prevailing. These methods adopt a two-tower framework to learn embedding vectors for query and item separately and thus leverage efficient approximate nearest neighbor (ANN) search to retrieve relevant items. However, existing EBR methods usually ignore inconsistent user behaviors in industrial multi-stage search systems, resulting in insufficient retrieval efficiency with a low commercial return. To tackle this challenge, we propose to improve EBR methods by learning Multi-level Multi-Grained Semantic Embeddings(MMSE). We propose the multi-stage information mining to exploit the ordered, clicked, unclicked and random sampled items in practical user behavior data, and then capture query-item similarity via a post-fusion strategy. We then propose multi-grained learning objectives that integrate the retrieval loss with global comparison ability and the ranking loss with local comparison ability to generate semantic embeddings. Both experiments on a real-world billion-scale dataset and online A/B tests verify the effectiveness of MMSE in achieving significant performance improvements on metrics such as offline recall and online conversion rate (CVR).

Data valuation is a growing research field that studies the influence of individual data points for machine learning (ML) models. Data Shapley, inspired by cooperative game theory and economics, is an effective method for data valuation. However, it is well-known that the Shapley value (SV) can be computationally expensive. Fortunately, Jia et al. (2019) showed that for K-Nearest Neighbors (KNN) models, the computation of Data Shapley is surprisingly simple and efficient. In this note, we revisit the work of Jia et al. (2019) and propose a more natural and interpretable utility function that better reflects the performance of KNN models. We derive the corresponding calculation procedure for the Data Shapley of KNN classifiers/regressors with the new utility functions. Our new approach, dubbed soft-label KNN-SV, achieves the same time complexity as the original method. We further provide an efficient approximation algorithm for soft-label KNN-SV based on locality sensitive hashing (LSH). Our experimental results demonstrate that Soft-label KNN-SV outperforms the original method on most datasets in the task of mislabeled data detection, making it a better baseline for future work on data valuation.

In this paper, we propose the Masked Space-Time Hash encoding (MSTH), a novel method for efficiently reconstructing dynamic 3D scenes from multi-view or monocular videos. Based on the observation that dynamic scenes often contain substantial static areas that result in redundancy in storage and computations, MSTH represents a dynamic scene as a weighted combination of a 3D hash encoding and a 4D hash encoding. The weights for the two components are represented by a learnable mask which is guided by an uncertainty-based objective to reflect the spatial and temporal importance of each 3D position. With this design, our method can reduce the hash collision rate by avoiding redundant queries and modifications on static areas, making it feasible to represent a large number of space-time voxels by hash tables with small size.Besides, without the requirements to fit the large numbers of temporally redundant features independently, our method is easier to optimize and converge rapidly with only twenty minutes of training for a 300-frame dynamic scene.As a result, MSTH obtains consistently better results than previous methods with only 20 minutes of training time and 130 MB of memory storage. Code is available at https://github.com/masked-spacetime-hashing/msth

To reduce the computational cost of convolutional neural networks (CNNs) for usage on resource-constrained devices, structured pruning approaches have shown promising results, drastically reducing floating-point operations (FLOPs) without substantial drops in accuracy. However, most recent methods require fine-tuning or specific training procedures to achieve a reasonable trade-off between retained accuracy and reduction in FLOPs. This introduces additional cost in the form of computational overhead and requires training data to be available. To this end, we propose HASTE (Hashing for Tractable Efficiency), a parameter-free and data-free module that acts as a plug-and-play replacement for any regular convolution module. It instantly reduces the network’s test-time inference cost without requiring any training or fine-tuning. We are able to drastically compress latent feature maps without sacrificing much accuracy by using locality-sensitive hashing (LSH) to detect redundancies in the channel dimension. Similar channels are aggregated to reduce the input and filter depth simultaneously, allowing for cheaper convolutions. We demonstrate our approach on the popular vision benchmarks CIFAR-10 and ImageNet. In particular, we are able to instantly drop 46.72% of FLOPs while only losing 1.25% accuracy by just swapping the convolution modules in a ResNet34 on CIFAR-10 for our HASTE module.

The shear number of sources that will be detected by next-generation radio surveys will be astronomical, which will result in serendipitous discoveries. Data-dependent deep hashing algorithms have been shown to be efficient at image retrieval tasks in the fields of computer vision and multimedia. However, there are limited applications of these methodologies in the field of astronomy. In this work, we utilize deep hashing to rapidly search for similar images in a large database. The experiment uses a balanced dataset of 2708 samples consisting of four classes: Compact, FRI, FRII, and Bent. The performance of the method was evaluated using the mean average precision (mAP) metric where a precision of 88.5\% was achieved. The experimental results demonstrate the capability to search and retrieve similar radio images efficiently and at scale. The retrieval is based on the Hamming distance between the binary hash of the query image and those of the reference images in the database.

Unsupervised hashing methods have attracted widespread attention with the explosive growth of large-scale data, which can greatly reduce storage and computation by learning compact binary codes. Existing unsupervised hashing methods attempt to exploit the valuable information from samples, which fails to take the local geometric structure of unlabeled samples into consideration. Moreover, hashing based on auto-encoders aims to minimize the reconstruction loss between the input data and binary codes, which ignores the potential consistency and complementarity of multiple sources data. To address the above issues, we propose a hashing algorithm based on auto-encoders for multi-view binary clustering, which dynamically learns affinity graphs with low-rank constraints and adopts collaboratively learning between auto-encoders and affinity graphs to learn a unified binary code, called Graph-Collaborated Auto-Encoder Hashing for Multi-view Binary Clustering (GCAE). Specifically, we propose a multi-view affinity graphs learning model with low-rank constraint, which can mine the underlying geometric information from multi-view data. Then, we design an encoder-decoder paradigm to collaborate the multiple affinity graphs, which can learn a unified binary code effectively. Notably, we impose the decorrelation and code balance constraints on binary codes to reduce the quantization errors. Finally, we utilize an alternating iterative optimization scheme to obtain the multi-view clustering results. Extensive experimental results on \(5\) public datasets are provided to reveal the effectiveness of the algorithm and its superior performance over other state-of-the-art alternatives.

Approximate nearest-neighbor search (ANNS) algorithms are a key part of the modern deep learning stack due to enabling efficient similarity search over high-dimensional vector space representations (i.e., embeddings) of data. Among various ANNS algorithms, graph-based algorithms are known to achieve the best throughput-recall tradeoffs. Despite the large scale of modern ANNS datasets, existing parallel graph based implementations suffer from significant challenges to scale to large datasets due to heavy use of locks and other sequential bottlenecks, which 1) prevents them from efficiently scaling to a large number of processors, and 2) results in nondeterminism that is undesirable in certain applications. In this paper, we introduce ParlayANN, a library of deterministic and parallel graph-based approximate nearest neighbor search algorithms, along with a set of useful tools for developing such algorithms. In this library, we develop novel parallel implementations for four state-of-the-art graph-based ANNS algorithms that scale to billion-scale datasets. Our algorithms are deterministic and achieve high scalability across a diverse set of challenging datasets. In addition to the new algorithmic ideas, we also conduct a detailed experimental study of our new algorithms as well as two existing non-graph approaches. Our experimental results both validate the effectiveness of our new techniques, and lead to a comprehensive comparison among ANNS algorithms on large scale datasets with a list of interesting findings.

Constructing a similarity graph from a set \(X\) of data points in \(\mathbb{R}^d\) is the first step of many modern clustering algorithms. However, typical constructions of a similarity graph have high time complexity, and a quadratic space dependency with respect to \(|X|\). We address this limitation and present a new algorithmic framework that constructs a sparse approximation of the fully connected similarity graph while preserving its cluster structure. Our presented algorithm is based on the kernel density estimation problem, and is applicable for arbitrary kernel functions. We compare our designed algorithm with the well-known implementations from the scikit-learn library and the FAISS library, and find that our method significantly outperforms the implementation from both libraries on a variety of datasets.

Anserini is a Lucene-based toolkit for reproducible information retrieval research in Java that has been gaining traction in the community. It provides retrieval capabilities for both “traditional” bag-of-words retrieval models such as BM25 as well as retrieval using learned sparse representations such as SPLADE. With Pyserini, which provides a Python interface to Anserini, users gain access to both sparse and dense retrieval models, as Pyserini implements bindings to the Faiss vector search library alongside Lucene inverted indexes in a uniform, consistent interface. Nevertheless, hybrid fusion techniques that integrate sparse and dense retrieval models need to stitch together results from two completely different “software stacks”, which creates unnecessary complexities and inefficiencies. However, the introduction of HNSW indexes for dense vector search in Lucene promises the integration of both dense and sparse retrieval within a single software framework. We explore exactly this integration in the context of Anserini. Experiments on the MS MARCO passage and BEIR datasets show that our Anserini HNSW integration supports (reasonably) effective and (reasonably) efficient approximate nearest neighbor search for dense retrieval models, using only Lucene.

In computational biology, \(k\)-mers and edit distance are fundamental concepts. However, little is known about the metric space of all \(k\)-mers equipped with the edit distance. In this work, we explore the structure of the \(k\)-mer space by studying its maximal independent sets (MISs). An MIS is a sparse sketch of all \(k\)-mers with nice theoretical properties, and therefore admits critical applications in clustering, indexing, hashing, and sketching large-scale sequencing data, particularly those with high error-rates. Finding an MIS is a challenging problem, as the size of a \(k\)-mer space grows geometrically with respect to \(k\). We propose three algorithms for this problem. The first and the most intuitive one uses a greedy strategy. The second method implements two techniques to avoid redundant comparisons by taking advantage of the locality-property of the \(k\)-mer space and the estimated bounds on the edit distance. The last algorithm avoids expensive calculations of the edit distance by translating the edit distance into the shortest path in a specifically designed graph. These algorithms are implemented and the calculated MISs of \(k\)-mer spaces and their statistical properties are reported and analyzed for \(k\) up to 15. Source code is freely available at https://github.com/Shao-Group/kmerspace .

Hash representation learning of multi-view heterogeneous data is the key to improving the accuracy of multimedia retrieval. However, existing methods utilize local similarity and fall short of deeply fusing the multi-view features, resulting in poor retrieval accuracy. Current methods only use local similarity to train their model. These methods ignore global similarity. Furthermore, most recent works fuse the multi-view features via a weighted sum or concatenation. We contend that these fusion methods are insufficient for capturing the interaction between various views. We present a novel Central Similarity Multi-View Hashing (CSMVH) method to address the mentioned problems. Central similarity learning is used for solving the local similarity problem, which can utilize the global similarity between the hash center and samples. We present copious empirical data demonstrating the superiority of gate-based fusion over conventional approaches. On the MS COCO and NUS-WIDE, the proposed CSMVH performs better than the state-of-the-art methods by a large margin (up to 11.41% mean Average Precision (mAP) improvement).

In recent years, hashing methods have been popular in the large-scale media search for low storage and strong representation capabilities. To describe objects with similar overall appearance but subtle differences, more and more studies focus on hashing-based fine-grained image retrieval. Existing hashing networks usually generate both local and global features through attention guidance on the same deep activation tensor, which limits the diversity of feature representations. To handle this limitation, we substitute convolutional descriptors for attention-guided features and propose an Attributes Grouping and Mining Hashing (AGMH), which groups and embeds the category-specific visual attributes in multiple descriptors to generate a comprehensive feature representation for efficient fine-grained image retrieval. Specifically, an Attention Dispersion Loss (ADL) is designed to force the descriptors to attend to various local regions and capture diverse subtle details. Moreover, we propose a Stepwise Interactive External Attention (SIEA) to mine critical attributes in each descriptor and construct correlations between fine-grained attributes and objects. The attention mechanism is dedicated to learning discrete attributes, which will not cost additional computations in hash codes generation. Finally, the compact binary codes are learned by preserving pairwise similarities. Experimental results demonstrate that AGMH consistently yields the best performance against state-of-the-art methods on fine-grained benchmark datasets.

Compressing videos into binary codes can improve retrieval speed and reduce storage overhead. However, learning accurate hash codes for video retrieval can be challenging due to high local redundancy and complex global dependencies between video frames, especially in the absence of labels. Existing self-supervised video hashing methods have been effective in designing expressive temporal encoders, but have not fully utilized the temporal dynamics and spatial appearance of videos due to less challenging and unreliable learning tasks. To address these challenges, we begin by utilizing the contrastive learning task to capture global spatio-temporal information of videos for hashing. With the aid of our designed augmentation strategies, which focus on spatial and temporal variations to create positive pairs, the learning framework can generate hash codes that are invariant to motion, scale, and viewpoint. Furthermore, we incorporate two collaborative learning tasks, i.e., frame order verification and scene change regularization, to capture local spatio-temporal details within video frames, thereby enhancing the perception of temporal structure and the modeling of spatio-temporal relationships. Our proposed Contrastive Hashing with Global-Local Spatio-temporal Information (CHAIN) outperforms state-of-the-art self-supervised video hashing methods on four video benchmark datasets. Our codes will be released.

To retrieve personalized campaigns and creatives while protecting user privacy, digital advertising is shifting from member-based identity to cohort-based identity. Under such identity regime, an accurate and efficient cohort building algorithm is desired to group users with similar characteristics. In this paper, we propose a scalable \(K\)-anonymous cohort building algorithm called {\em consecutive consistent weighted sampling} (CCWS). The proposed method combines the spirit of the (\(p\)-powered) consistent weighted sampling and hierarchical clustering, so that the \(K\)-anonymity is ensured by enforcing a lower bound on the size of cohorts. Evaluations on a LinkedIn dataset consisting of \(>70\)M users and ads campaigns demonstrate that CCWS achieves substantial improvements over several hashing-based methods including sign random projections (SignRP), minwise hashing (MinHash), as well as the vanilla CWS.

The multi-view hash method converts heterogeneous data from multiple views into binary hash codes, which is one of the critical technologies in multimedia retrieval. However, the current methods mainly explore the complementarity among multiple views while lacking confidence learning and fusion. Moreover, in practical application scenarios, the single-view data contain redundant noise. To conduct the confidence learning and eliminate unnecessary noise, we propose a novel Adaptive Confidence Multi-View Hashing (ACMVH) method. First, a confidence network is developed to extract useful information from various single-view features and remove noise information. Furthermore, an adaptive confidence multi-view network is employed to measure the confidence of each view and then fuse multi-view features through a weighted summation. Lastly, a dilation network is designed to further enhance the feature representation of the fused features. To the best of our knowledge, we pioneer the application of confidence learning into the field of multimedia retrieval. Extensive experiments on two public datasets show that the proposed ACMVH performs better than state-of-the-art methods (maximum increase of 3.24%). The source code is available at https://github.com/HackerHyper/ACMVH.

The multi-modal hashing method is widely used in multimedia retrieval. It can fuse multi-source data to generate binary hash code. However, the current multi-modal methods have the problem of low retrieval accuracy. The reason is that the individual backbone networks have limited feature expression capabilities and are not jointly pre-trained on large-scale unsupervised multi-modal data. To solve this problem, we propose a new baseline CLIP Multi-modal Hashing (CLIPMH) method. It uses CLIP model to extract text and image features, and then fuse to generate hash code. CLIP improves the expressiveness of each modal feature. In this way, it can greatly improve the retrieval performance of multi-modal hashing methods. In comparison to state-of-the-art unsupervised and supervised multi-modal hashing methods, experiments reveal that the proposed CLIPMH can significantly enhance performance (Maximum increase of 8.38%). CLIP also has great advantages over the text and visual backbone networks commonly used before.

Our work focuses on tackling large-scale fine-grained image retrieval as ranking the images depicting the concept of interests (i.e., the same sub-category labels) highest based on the fine-grained details in the query. It is desirable to alleviate the challenges of both fine-grained nature of small inter-class variations with large intra-class variations and explosive growth of fine-grained data for such a practical task. In this paper, we propose attribute-aware hashing networks with self-consistency for generating attribute-aware hash codes to not only make the retrieval process efficient, but also establish explicit correspondences between hash codes and visual attributes. Specifically, based on the captured visual representations by attention, we develop an encoder-decoder structure network of a reconstruction task to unsupervisedly distill high-level attribute-specific vectors from the appearance-specific visual representations without attribute annotations. Our models are also equipped with a feature decorrelation constraint upon these attribute vectors to strengthen their representative abilities. Then, driven by preserving original entities’ similarity, the required hash codes can be generated from these attribute-specific vectors and thus become attribute-aware. Furthermore, to combat simplicity bias in deep hashing, we consider the model design from the perspective of the self-consistency principle and propose to further enhance models’ self-consistency by equipping an additional image reconstruction path. Comprehensive quantitative experiments under diverse empirical settings on six fine-grained retrieval datasets and two generic retrieval datasets show the superiority of our models over competing methods.

Cayley hash functions are cryptographic hashes constructed from Cayley graphs of groups. The hash function proposed by Shpilrain and Sosnovski (2016), based on linear functions over a finite field, was proven insecure. This paper shows that the proposal by Ghaffari and Mostaghim (2018) that uses the Shpilrain and Sosnovski’s hash in its construction is also insecure. We demonstrate its security vulnerability by constructing collisions.

Recommender systems have become an essential component of many online platforms, providing personalized recommendations to users. A crucial aspect is embedding techniques that coverts the high-dimensional discrete features, such as user and item IDs, into low-dimensional continuous vectors and can enhance the recommendation performance. Applying embedding techniques captures complex entity relationships and has spurred substantial research. In this survey, we provide an overview of the recent literature on embedding techniques in recommender systems. This survey covers embedding methods like collaborative filtering, self-supervised learning, and graph-based techniques. Collaborative filtering generates embeddings capturing user-item preferences, excelling in sparse data. Self-supervised methods leverage contrastive or generative learning for various tasks. Graph-based techniques like node2vec exploit complex relationships in network-rich environments. Addressing the scalability challenges inherent to embedding methods, our survey delves into innovative directions within the field of recommendation systems. These directions aim to enhance performance and reduce computational complexity, paving the way for improved recommender systems. Among these innovative approaches, we will introduce Auto Machine Learning (AutoML), hash techniques, and quantization techniques in this survey. We discuss various architectures and techniques and highlight the challenges and future directions in these aspects. This survey aims to provide a comprehensive overview of the state-of-the-art in this rapidly evolving field and serve as a useful resource for researchers and practitioners working in the area of recommender systems.

Category information plays a crucial role in enhancing the quality and personalization of recommender systems. Nevertheless, the availability of item category information is not consistently present, particularly in the context of ID-based recommendations. In this work, we propose a novel approach to automatically learn and generate entity (i.e., user or item) category trees for ID-based recommendation. Specifically, we devise a differentiable vector quantization framework for automatic category tree generation, namely CAGE, which enables the simultaneous learning and refinement of categorical code representations and entity embeddings in an end-to-end manner, starting from the randomly initialized states. With its high adaptability, CAGE can be easily integrated into both sequential and non-sequential recommender systems. We validate the effectiveness of CAGE on various recommendation tasks including list completion, collaborative filtering, and click-through rate prediction, across different recommendation models. We release the code and data for others to reproduce the reported results.

An efficient solution to the large-scale recommender system is to represent users and items as binary hash codes in the Hamming space. Towards this end, existing methods tend to code users by modeling their Hamming similarities with the items they historically interact with, which are termed as the first-order similarities in this work. Despite their efficiency, these methods suffer from the suboptimal representative capacity, since they forgo the correlation established by connecting multiple first-order similarities, i.e., the relation among the indirect instances, which could be defined as the high-order similarity. To tackle this drawback, we propose to model both the first- and the high-order similarities in the Hamming space through the user-item bipartite graph. Therefore, we develop a novel learning to hash framework, namely Hamming Spatial Graph Convolutional Networks (HS-GCN), which explicitly models the Hamming similarity and embeds it into the codes of users and items. Extensive experiments on three public benchmark datasets demonstrate that our proposed model significantly outperforms several state-of-the-art hashing models, and obtains performance comparable with the real-valued recommendation models.

Unsupervised hashing has received extensive research focus on the past decade, which typically aims at preserving a predefined metric (i.e. Euclidean metric) in the Hamming space. To this end, the encoding functions of the existing hashing are typically quasi-isometric, which devote to reducing the quantization loss from the target metric space to the discrete Hamming space. However, it is indeed problematic to directly minimize such error, since such mentioned two metric spaces are heterogeneous, and the quasi-isometric mapping is non-linear. The former leads to inconsistent feature distributions, while the latter leads to problematic optimization issues. In this paper, we propose a novel unsupervised hashing method, termed Sparsity-Induced Generative Adversarial Hashing (SiGAH), to encode large-scale high-dimensional features into binary codes, which well solves the two problems through a generative adversarial training framework. Instead of minimizing the quantization loss, our key innovation lies in enforcing the learned Hamming space to have similar data distribution to the target metric space via a generative model. In particular, we formulate a ReLU-based neural network as a generator to output binary codes and an MSE-loss based auto-encoder network as a discriminator, upon which a generative adversarial learning is carried out to train hash functions. Furthermore, to generate the synthetic features from the hash codes, a compressed sensing procedure is introduced into the generative model, which enforces the reconstruction boundary of binary codes to be consistent with that of original features. Finally, such generative adversarial framework can be trained via the Adam optimizer. Experimental results on four benchmarks, i.e., Tiny100K, GIST1M, Deep1M, and MNIST, have shown that the proposed SiGAH has superior performance over the state-of-the-art approaches.

Deep hashing has been extensively applied to massive image retrieval due to its efficiency and effectiveness. Recently, several adversarial attacks have been presented to reveal the vulnerability of deep hashing models against adversarial examples. However, existing attack methods suffer from degraded performance or inefficiency because they underutilize the semantic relations between original samples or spend a lot of time learning these relations with a deep neural network. In this paper, we propose a novel Pharos-guided Attack, dubbed PgA, to evaluate the adversarial robustness of deep hashing networks reliably and efficiently. Specifically, we design pharos code to represent the semantics of the benign image, which preserves the similarity to semantically relevant samples and dissimilarity to irrelevant ones. It is proven that we can quickly calculate the pharos code via a simple math formula. Accordingly, PgA can directly conduct a reliable and efficient attack on deep hashing-based retrieval by maximizing the similarity between the hash code of the adversarial example and the pharos code. Extensive experiments on the benchmark datasets verify that the proposed algorithm outperforms the prior state-of-the-arts in both attack strength and speed.

Learning the hash representation of multi-view heterogeneous data is an important task in multimedia retrieval. However, existing methods fail to effectively fuse the multi-view features and utilize the metric information provided by the dissimilar samples, leading to limited retrieval precision. Current methods utilize weighted sum or concatenation to fuse the multi-view features. We argue that these fusion methods cannot capture the interaction among different views. Furthermore, these methods ignored the information provided by the dissimilar samples. We propose a novel deep metric multi-view hashing (DMMVH) method to address the mentioned problems. Extensive empirical evidence is presented to show that gate-based fusion is better than typical methods. We introduce deep metric learning to the multi-view hashing problems, which can utilize metric information of dissimilar samples. On the MIR-Flickr25K, MS COCO, and NUS-WIDE, our method outperforms the current state-of-the-art methods by a large margin (up to 15.28 mean Average Precision (mAP) improvement).

The asymptotic rate vs. distance problem is a long-standing fundamental problem in coding theory. The best upper bound to date was given in 1977 and has received since then numerous proofs and interpretations. Here we provide a new, elementary proof of this bound based on counting walks in the Hamming cube.

Binary code similarity detection is to detect the similarity of code at binary (assembly) level without source code. Existing works have their limitations when dealing with mutated binary code generated by different compiling options. In this paper, we propose a novel approach to addressing this problem. By inspecting the binary code, we found that generally, within a function, some instructions aim to calculate (prepare) values for other instructions. The latter instructions are defined by us as key instructions. Currently, we define four categories of key instructions: calling subfunctions, comparing instruction, returning instruction, and memory-store instruction. Thus if we symbolically execute similar binary codes, symbolic values at these key instructions are expected to be similar. As such, we implement a prototype tool, which has three steps. First, it symbolically executes binary code; Second, it extracts symbolic values at defined key instructions into a graph; Last, it compares the symbolic graph similarity. In our implementation, we also address some problems, including path explosion and loop handling.

Few-shot image classification is the task of classifying unseen images to one of N mutually exclusive classes, using only a small number of training examples for each class. The limited availability of these examples (denoted as K) presents a significant challenge to classification accuracy in some cases. To address this, we have developed a method for augmenting the set of K with an addition set of A retrieved images. We call this system Retrieval-Augmented Few-shot Image Classification (RAFIC). Through a series of experiments, we demonstrate that RAFIC markedly improves performance of few-shot image classification across two challenging datasets. RAFIC consists of two main components: (a) a retrieval component which uses CLIP, LAION-5B, and faiss, in order to efficiently retrieve images similar to the supplied images, and (b) retrieval meta-learning, which learns to judiciously utilize the retrieved images. Code and data is available at github.com/amirziai/rafic.

Latent factor models are the dominant backbones of contemporary recommender systems (RSs) given their performance advantages, where a unique vector embedding with a fixed dimensionality (e.g., 128) is required to represent each entity (commonly a user/item). Due to the large number of users and items on e-commerce sites, the embedding table is arguably the least memory-efficient component of RSs. For any lightweight recommender that aims to efficiently scale with the growing size of users/items or to remain applicable in resource-constrained settings, existing solutions either reduce the number of embeddings needed via hashing, or sparsify the full embedding table to switch off selected embedding dimensions. However, as hash collision arises or embeddings become overly sparse, especially when adapting to a tighter memory budget, those lightweight recommenders inevitably have to compromise their accuracy. To this end, we propose a novel compact embedding framework for RSs, namely Compositional Embedding with Regularized Pruning (CERP). Specifically, CERP represents each entity by combining a pair of embeddings from two independent, substantially smaller meta-embedding tables, which are then jointly pruned via a learnable element-wise threshold. In addition, we innovatively design a regularized pruning mechanism in CERP, such that the two sparsified meta-embedding tables are encouraged to encode information that is mutually complementary. Given the compatibility with agnostic latent factor models, we pair CERP with two popular recommendation models for extensive experiments, where results on two real-world datasets under different memory budgets demonstrate its superiority against state-of-the-art baselines. The codebase of CERP is available in https://github.com/xurong-liang/CERP.

Sparse data are common. The traditional handcrafted'' features are often sparse. Embedding vectors from trained models can also be very sparse, for example, embeddings trained via theReLu’’ activation function. In this paper, we report our exploration of efficient search in sparse data with graph-based ANN algorithms (e.g., HNSW, or SONG which is the GPU version of HNSW), which are popular in industrial practice, e.g., search and ads (advertising). We experiment with the proprietary ads targeting application, as well as benchmark public datasets. For ads targeting, we train embeddings with the standard cosine two-tower'' model and we also develop thechi-square two-tower’’ model. Both models produce (highly) sparse embeddings when they are integrated with the ReLu'' activation function. In EBR (embedding-based retrieval) applications, after we the embeddings are trained, the next crucial task is the approximate near neighbor (ANN) search for serving. While there are many ANN algorithms we can choose from, in this study, we focus on the graph-based ANN algorithm (e.g., HNSW-type). Sparse embeddings should help improve the efficiency of EBR. One benefit is the reduced memory cost for the embeddings. The other obvious benefit is the reduced computational time for evaluating similarities, because, for graph-based ANN algorithms such as HNSW, computing similarities is often the dominating cost. In addition to the effort on leveraging data sparsity for storage and computation, we also integratesign cauchy random projections’’ (SignCRP) to hash vectors to bits, to further reduce the memory cost and speed up the ANN search. In NIPS’13, SignCRP was proposed to hash the chi-square similarity, which is a well-adopted nonlinear kernel in NLP and computer vision. Therefore, the chi-square two-tower model, SignCRP, and HNSW are now tightly integrated.

Hashing is very popular for remote sensing image search. This article proposes a multiview hashing with learnable parameters to retrieve the queried images for a large-scale remote sensing dataset. Existing methods always neglect that real-world remote sensing data lies on a low-dimensional manifold embedded in high-dimensional ambient space. Unlike previous methods, this article proposes to learn the consensus compact codes in a view-specific low-dimensional subspace. Furthermore, we have added a hyperparameter learnable module to avoid complex parameter tuning. In order to prove the effectiveness of our method, we carried out experiments on three widely used remote sensing data sets and compared them with seven state-of-the-art methods. Extensive experiments show that the proposed method can achieve competitive results compared to the other method.

Unsupervised video hashing usually optimizes binary codes by learning to reconstruct input videos. Such reconstruction constraint spends much effort on frame-level temporal context changes without focusing on video-level global semantics that are more useful for retrieval. Hence, we address this problem by decomposing video information into reconstruction-dependent and semantic-dependent information, which disentangles the semantic extraction from reconstruction constraint. Specifically, we first design a simple dual-stream structure, including a temporal layer and a hash layer. Then, with the help of semantic similarity knowledge obtained from self-supervision, the hash layer learns to capture information for semantic retrieval, while the temporal layer learns to capture the information for reconstruction. In this way, the model naturally preserves the disentangled semantics into binary codes. Validated by comprehensive experiments, our method consistently outperforms the state-of-the-arts on three video benchmarks.

Many hashing algorithms including minwise hashing (MinHash), one permutation hashing (OPH), and consistent weighted sampling (CWS) generate integers of \(B\) bits. With \(k\) hashes for each data vector, the storage would be \(B\times k\) bits; and when used for large-scale learning, the model size would be \(2^B\times k\), which can be expensive. A standard strategy is to use only the lowest \(b\) bits out of the \(B\) bits and somewhat increase \(k\), the number of hashes. In this study, we propose to re-use the hashes by partitioning the \(B\) bits into \(m\) chunks, e.g., \(b\times m =B\). Correspondingly, the model size becomes \(m\times 2^b \times k\), which can be substantially smaller than the original \(2^B\times k\). Our theoretical analysis reveals that by partitioning the hash values into \(m\) chunks, the accuracy would drop. In other words, using \(m\) chunks of \(B/m\) bits would not be as accurate as directly using \(B\) bits. This is due to the correlation from re-using the same hash. On the other hand, our analysis also shows that the accuracy would not drop much for (e.g.,) \(m=2\sim 4\). In some regions, Pb-Hash still works well even for \(m\) much larger than 4. We expect Pb-Hash would be a good addition to the family of hashing methods/applications and benefit industrial practitioners. We verify the effectiveness of Pb-Hash in machine learning tasks, for linear SVM models as well as deep learning models. Since the hashed data are essentially categorical (ID) features, we follow the standard practice of using embedding tables for each hash. With Pb-Hash, we need to design an effective strategy to combine \(m\) embeddings. Our study provides an empirical evaluation on four pooling schemes: concatenation, max pooling, mean pooling, and product pooling. There is no definite answer which pooling would be always better and we leave that for future study.

With the rapid development of GPU (Graphics Processing Unit) technologies and neural networks, we can explore more appropriate data structures and algorithms. Recent progress shows that neural networks can partly replace traditional data structures. In this paper, we proposed a novel DNN (Deep Neural Network)-based learned locality-sensitive hashing, called LLSH, to efficiently and flexibly map high-dimensional data to low-dimensional space. LLSH replaces the traditional LSH (Locality-sensitive Hashing) function families with parallel multi-layer neural networks, which reduces the time and memory consumption and guarantees query accuracy simultaneously. The proposed LLSH demonstrate the feasibility of replacing the hash index with learning-based neural networks and open a new door for developers to design and configure data organization more accurately to improve information-searching performance. Extensive experiments on different types of datasets show the superiority of the proposed method in query accuracy, time consumption, and memory usage.

Minwise hashing (MinHash) is a standard algorithm widely used in the industry, for large-scale search and learning applications with the binary (0/1) Jaccard similarity. One common use of MinHash is for processing massive n-gram text representations so that practitioners do not have to materialize the original data (which would be prohibitive). Another popular use of MinHash is for building hash tables to enable sub-linear time approximate near neighbor (ANN) search. MinHash has also been used as a tool for building large-scale machine learning systems. The standard implementation of MinHash requires applying \(K\) random permutations. In comparison, the method of one permutation hashing (OPH), is an efficient alternative of MinHash which splits the data vectors into \(K\) bins and generates hash values within each bin. OPH is substantially more efficient and also more convenient to use. In this paper, we combine the differential privacy (DP) with OPH (as well as MinHash), to propose the DP-OPH framework with three variants: DP-OPH-fix, DP-OPH-re and DP-OPH-rand, depending on which densification strategy is adopted to deal with empty bins in OPH. A detailed roadmap to the algorithm design is presented along with the privacy analysis. An analytical comparison of our proposed DP-OPH methods with the DP minwise hashing (DP-MH) is provided to justify the advantage of DP-OPH. Experiments on similarity search confirm the merits of DP-OPH, and guide the choice of the proper variant in different practical scenarios. Our technique is also extended to bin-wise consistent weighted sampling (BCWS) to develop a new DP algorithm called DP-BCWS for non-binary data. Experiments on classification tasks demonstrate that DP-BCWS is able to achieve excellent utility at around \(\epsilon = 5\sim 10\), where \(\epsilon\) is the standard parameter in the language of \((\epsilon, \delta)\)-DP.

This paper introduces a new challenge for image similarity search in the context of fashion, addressing the inherent ambiguity in this domain stemming from complex images. We present Referred Visual Search (RVS), a task allowing users to define more precisely the desired similarity, following recent interest in the industry. We release a new large public dataset, LRVS-Fashion, consisting of 272k fashion products with 842k images extracted from fashion catalogs, designed explicitly for this task. However, unlike traditional visual search methods in the industry, we demonstrate that superior performance can be achieved by bypassing explicit object detection and adopting weakly-supervised conditional contrastive learning on image tuples. Our method is lightweight and demonstrates robustness, reaching Recall at one superior to strong detection-based baselines against 2M distractors. The dataset is available at https://huggingface.co/datasets/Slep/LAION-RVS-Fashion .

Consistent hashing is a technique that can minimize key remapping when the number of hash buckets changes. The paper proposes a fast consistent hash algorithm (called power consistent hash) that has \(O(1)\) expected time for key lookup, independent of the number of buckets. Hash values are computed in real time. No search data structure is constructed to store bucket ranges or key mappings. The algorithm has a lightweight design using \(O(1)\) space with superior scalability. In particular, it uses two auxiliary hash functions to achieve distribution uniformity and \(O(1)\) expected time for key lookup. Furthermore, it performs consistent hashing such that only a minimal number of keys are remapped when the number of buckets changes. Consistent hashing has a wide range of use cases, including load balancing, distributed caching, and distributed key-value stores. The proposed algorithm is faster than well-known consistent hash algorithms with \(O(log n)\) lookup time.

A minimal perfect hash function (MPHF) maps a set S of n keys to the first n integers without collisions. There is a lower bound of n*log(e)=1.44n bits needed to represent an MPHF. This can be reached by a brute-force algorithm that tries e^n hash function seeds in expectation and stores the first seed leading to an MPHF. The most space-efficient previous algorithms for constructing MPHFs all use such a brute-force approach as a basic building block. In this paper, we introduce ShockHash - Small, heavily overloaded cuckoo hash tables for minimal perfect hashing. ShockHash uses two hash functions h_0 and h_1, hoping for the existence of a function f : S->{0, 1} such that x -> h_{f(x)}(x) is an MPHF on S. It then uses a 1-bit retrieval data structure to store f using n + o(n) bits. In graph terminology, ShockHash generates n-edge random graphs until stumbling on a pseudoforest - where each component contains as many edges as nodes. Using cuckoo hashing, ShockHash then derives an MPHF from the pseudoforest in linear time. We show that ShockHash needs to try only about (e/2)^n=1.359^n seeds in expectation. This reduces the space for storing the seed by roughly n bits (maintaining the asymptotically optimal space consumption) and speeds up construction by almost a factor of 2^n compared to brute-force. Bipartite ShockHash reduces the expected construction time again to 1.166^n by maintaining a pool of candidate hash functions and checking all possible pairs. ShockHash as a building block within the RecSplit framework can be constructed up to 3 orders of magnitude faster than competing approaches. It can build an MPHF for 10 million keys with 1.489 bits per key in about half an hour. When instead using ShockHash after an efficient k-perfect hash function, it achieves space usage similar to the best competitors, while being significantly faster to construct and query.

String matching is the problem of finding all the occurrences of a pattern in a text. We propose improved versions of the fast family of string matching algorithms based on hashing \(q\)-grams. The improvement consists of considering minimal values \(q\) such that each \(q\)-grams of the pattern has a unique hash value. The new algorithms are fastest than algorithm of the HASH family for short patterns on large size alphabets.

We present {\bf Sli}ding Blo{\bf ck} Hashing (Slick), a simple hash table data structure that combines high performance with very good space efficiency. This preliminary report outlines avenues for analysis and implementation that we intend to pursue.

We present ElasticHash, a novel approach for high-quality, efficient, and large-scale semantic image similarity search. It is based on a deep hashing model to learn hash codes for fine-grained image similarity search in natural images and a two-stage method for efficiently searching binary hash codes using Elasticsearch (ES). In the first stage, a coarse search based on short hash codes is performed using multi-index hashing and ES terms lookup of neighboring hash codes. In the second stage, the list of results is re-ranked by computing the Hamming distance on long hash codes. We evaluate the retrieval performance of \textit{ElasticHash} for more than 120,000 query images on about 6.9 million database images of the OpenImages data set. The results show that our approach achieves high-quality retrieval results and low search latencies.

This paper focuses on a class of binary orthogonal optimization problems frequently arising in semantic hashing. Consider that this class of problems may have an empty feasible set, rendering them not well-defined. We introduce an equivalent model involving a restricted Stiefel manifold and a matrix box set, and then investigate its penalty problems induced by the \(\ell_1\)-distance from the box set and its Moreau envelope. The two penalty problems are always well-defined. Moreover, they serve as the global exact penalties provided that the original feasible set is non-empty. Notably, the penalty problem induced by the Moreau envelope is a smooth optimization over an embedded submanifold with a favorable structure. We develop a retraction-based line-search Riemannian gradient method to address the penalty problem. Finally, the proposed method is applied to supervised and unsupervised hashing tasks and is compared with several popular methods on the MNIST and CIFAR-10 datasets. The numerical comparisons reveal that our algorithm is significantly superior to other solvers in terms of feasibility violation, and it is comparable even superior to others in terms of evaluation metrics related to the Hamming distance.

When investigating a malicious file, searching for related files is a common task that malware analysts must perform. Given that production malware corpora may contain over a billion files and consume petabytes of storage, many feature extraction and similarity search approaches are computationally infeasible. Our work explores the potential of antivirus (AV) scan data as a scalable source of features for malware. This is possible because AV scan reports are widely available through services such as VirusTotal and are ~100x smaller than the average malware sample. The information within an AV scan report is abundant with information and can indicate a malicious file’s family, behavior, target operating system, and many other characteristics. We introduce AVScan2Vec, a language model trained to comprehend the semantics of AV scan data. AVScan2Vec ingests AV scan data for a malicious file and outputs a meaningful vector representation. AVScan2Vec vectors are ~3 to 85x smaller than popular alternatives in use today, enabling faster vector comparisons and lower memory usage. By incorporating Dynamic Continuous Indexing, we show that nearest-neighbor queries on AVScan2Vec vectors can scale to even the largest malware production datasets. We also demonstrate that AVScan2Vec vectors are superior to other leading malware feature vector representations across nearly all classification, clustering, and nearest-neighbor lookup algorithms that we evaluated.

Deep learning-based recommendation systems (e.g., DLRMs) are widely used AI models to provide high-quality personalized recommendations. Training data used for modern recommendation systems commonly includes categorical features taking on tens-of-millions of possible distinct values. These categorical tokens are typically assigned learned vector representations, that are stored in large embedding tables, on the order of 100s of GB. Storing and accessing these tables represent a substantial burden in commercial deployments. Our work proposes MEM-REC, a novel alternative representation approach for embedding tables. MEM-REC leverages bloom filters and hashing methods to encode categorical features using two cache-friendly embedding tables. The first table (token embedding) contains raw embeddings (i.e. learned vector representation), and the second table (weight embedding), which is much smaller, contains weights to scale these raw embeddings to provide better discriminative capability to each data point. We provide a detailed architecture, design and analysis of MEM-REC addressing trade-offs in accuracy and computation requirements, in comparison with state-of-the-art techniques. We show that MEM-REC can not only maintain the recommendation quality and significantly reduce the memory footprint for commercial scale recommendation models but can also improve the embedding latency. In particular, based on our results, MEM-REC compresses the MLPerf CriteoTB benchmark DLRM model size by 2900x and performs up to 3.4x faster embeddings while achieving the same AUC as that of the full uncompressed model.

The vulnerability in the algorithm supply chain of deep learning has imposed new challenges to image retrieval systems in the downstream. Among a variety of techniques, deep hashing is gaining popularity. As it inherits the algorithmic backend from deep learning, a handful of attacks are recently proposed to disrupt normal image retrieval. Unfortunately, the defense strategies in softmax classification are not readily available to be applied in the image retrieval domain. In this paper, we propose an efficient and unsupervised scheme to identify unique adversarial behaviors in the hamming space. In particular, we design three criteria from the perspectives of hamming distance, quantization loss and denoising to defend against both untargeted and targeted attacks, which collectively limit the adversarial space. The extensive experiments on four datasets demonstrate 2-23% improvements of detection rates with minimum computational overhead for real-time image queries.

Graph-based approaches to nearest neighbor search are popular and powerful tools for handling large datasets in practice, but they have limited theoretical guarantees. We study the worst-case performance of recent graph-based approximate nearest neighbor search algorithms, such as HNSW, NSG and DiskANN. For DiskANN, we show that its “slow preprocessing” version provably supports approximate nearest neighbor search query with constant approximation ratio and poly-logarithmic query time, on data sets with bounded “intrinsic” dimension. For the other data structure variants studied, including DiskANN with “fast preprocessing”, HNSW and NSG, we present a family of instances on which the empirical query time required to achieve a “reasonable” accuracy is linear in instance size. For example, for DiskANN, we show that the query procedure can take at least \(0.1 n\) steps on instances of size \(n\) before it encounters any of the \(5\) nearest neighbors of the query.

Finding the nearest neighbor to a hyperplane (or Point-to-Hyperplane Nearest Neighbor Search, simply P2HNNS) is a new and challenging problem with applications in many research domains. While existing state-of-the-art hashing schemes (e.g., NH and FH) are able to achieve sublinear time complexity without the assumption of the data being in a unit hypersphere, they require an asymmetric transformation, which increases the data dimension from \(d\) to \(Ω(d^2)\). This leads to considerable overhead for indexing and incurs significant distortion errors. In this paper, we investigate a tree-based approach for solving P2HNNS using the classical Ball-Tree index. Compared to hashing-based methods, tree-based methods usually require roughly linear costs for construction, and they provide different kinds of approximations with excellent flexibility. A simple branch-and-bound algorithm with a novel lower bound is first developed on Ball-Tree for performing P2HNNS. Then, a new tree structure named BC-Tree, which maintains the Ball and Cone structures in the leaf nodes of Ball-Tree, is described together with two effective strategies, i.e., point-level pruning and collaborative inner product computing. BC-Tree inherits both the low construction cost and lightweight property of Ball-Tree while providing a similar or more efficient search. Experimental results over 16 real-world data sets show that Ball-Tree and BC-Tree are around 1.1\(\sim\)10\(\times\) faster than NH and FH, and they can reduce the index size and indexing time by about 1\(\sim\)3 orders of magnitudes on average. The code is available at \url{https://github.com/HuangQiang/BC-Tree}.

Existing watermarking algorithms are vulnerable to paraphrase attacks because of their token-level design. To address this issue, we propose SemStamp, a robust sentence-level semantic watermarking algorithm based on locality-sensitive hashing (LSH), which partitions the semantic space of sentences. The algorithm encodes and LSH-hashes a candidate sentence generated by an LLM, and conducts sentence-level rejection sampling until the sampled sentence falls in watermarked partitions in the semantic embedding space. A margin-based constraint is used to enhance its robustness. To show the advantages of our algorithm, we propose a “bigram” paraphrase attack using the paraphrase that has the fewest bigram overlaps with the original sentence. This attack is shown to be effective against the existing token-level watermarking method. Experimental results show that our novel semantic watermark algorithm is not only more robust than the previous state-of-the-art method on both common and bigram paraphrase attacks, but also is better at preserving the quality of generation.

The Sparse Johnson-Lindenstrauss Transform of Kane and Nelson (SODA 2012) provides a linear dimensionality-reducing map \(A \in \mathbb{R}^{m \times u}\) in \(ℓ₂\) that preserves distances up to distortion of \(1 + \epsilon\) with probability \(1 - \delta\), where \(m = O(\epsilon^{-2} log 1/\delta)\) and each column of \(A\) has \(O(\epsilon m)\) non-zero entries. The previous analyses of the Sparse Johnson-Lindenstrauss Transform all assumed access to a \(Ω(log 1/\delta)\)-wise independent hash function. The main contribution of this paper is a more general analysis of the Sparse Johnson-Lindenstrauss Transform with less assumptions on the hash function. We also show that the Mixed Tabulation hash function of Dahlgaard, Knudsen, Rotenberg, and Thorup (FOCS 2015) satisfies the conditions of our analysis, thus giving us the first analysis of a Sparse Johnson-Lindenstrauss Transform that works with a practical hash function.

Typically, metadata of images are stored in a specific data segment of the image file. However, to securely detect changes, data can also be embedded within images. This follows the goal to invisibly and robustly embed as much information as possible to, ideally, even survive compression. This work searches for embedding principles which allow to distinguish between unintended changes by lossy image compression and malicious manipulation of the embedded message based on the change of its perceptual or robust hash. Different embedding and compression algorithms are compared. The study shows that embedding a message via integer wavelet transform and compression with Karhunen-Loeve-transform yields the best results. However, it was not possible to distinguish between manipulation and compression in all cases.

With the explosive growth in the number of fine-grained images in the Internet era, it has become a challenging problem to perform fast and efficient retrieval from large-scale fine-grained images. Among the many retrieval methods, hashing methods are widely used due to their high efficiency and small storage space occupation. Fine-grained hashing is more challenging than traditional hashing problems due to the difficulties such as low inter-class variances and high intra-class variances caused by the characteristics of fine-grained images. To improve the retrieval accuracy of fine-grained hashing, we propose a cascaded network to learn compact and highly semantic hash codes, and introduce an attention-guided data augmentation method. We refer to this network as a cascaded hierarchical data augmentation network. We also propose a novel approach to coordinately balance the loss of multi-task learning. We do extensive experiments on some common fine-grained visual classification datasets. The experimental results demonstrate that our proposed method outperforms several state-of-art hashing methods and can effectively improve the accuracy of fine-grained retrieval. The source code is publicly available: https://github.com/kaiba007/FG-CNET.

Hashing methods have made significant progress in cross-modal retrieval tasks with fast query speed and low storage cost. Among them, deep learning-based hashing achieves better performance on large-scale data due to its excellent extraction and representation ability for nonlinear heterogeneous features. However, there are still two main challenges in catastrophic forgetting when data with new categories arrive continuously, and time-consuming for non-continuous hashing retrieval to retrain for updating. To this end, we, in this paper, propose a novel deep lifelong cross-modal hashing to achieve lifelong hashing retrieval instead of re-training hash function repeatedly when new data arrive. Specifically, we design lifelong learning strategy to update hash functions by directly training the incremental data instead of retraining new hash functions using all the accumulated data, which significantly reduce training time. Then, we propose lifelong hashing loss to enable original hash codes participate in lifelong learning but remain invariant, and further preserve the similarity and dis-similarity among original and incremental hash codes to maintain performance. Additionally, considering distribution heterogeneity when new data arriving continuously, we introduce multi-label semantic similarity to supervise hash learning, and it has been proven that the similarity improves performance with detailed analysis. Experimental results on benchmark datasets show that the proposed methods achieves comparative performance comparing with recent state-of-the-art cross-modal hashing methods, and it yields substantial average increments over 20\% in retrieval accuracy and almost reduces over 80\% training time when new data arrives continuously.

We introduce and analyse a family of hash and predicate functions that are more likely to produce collisions for small reducible configurations of vectors. These may offer practical improvements to lattice sieving for short vectors. In particular, in one asymptotic regime the family exhibits significantly different convergent behaviour than existing hash functions and predicates.

Training and inference on edge devices often requires an efficient setup due to computational limitations. While pre-computing data representations and caching them on a server can mitigate extensive edge device computation, this leads to two challenges. First, the amount of storage required on the server that scales linearly with the number of instances. Second, the bandwidth required to send extensively large amounts of data to an edge device. To reduce the memory footprint of pre-computed data representations, we propose a simple, yet effective approach that uses randomly initialized hyperplane projections. To further reduce their size by up to 98.96%, we quantize the resulting floating-point representations into binary vectors. Despite the greatly reduced size, we show that the embeddings remain effective for training models across various English and German sentence classification tasks that retain 94%–99% of their floating-point.

Template matching is a fundamental problem in computer vision with applications in fields including object detection, image registration, and object tracking. Current methods rely on nearest-neighbour (NN) matching, where the query feature space is converted to NN space by representing each query pixel with its NN in the template. NN-based methods have been shown to perform better in occlusions, appearance changes, and non-rigid transformations; however, they scale poorly with high-resolution data and high feature dimensions. We present an NN-based method which efficiently reduces the NN computations and introduces filtering in the NN fields (NNFs). A vector quantization step is introduced before the NN calculation to represent the template with \(k\) features, and the filter response over the NNFs is used to compare the template and query distributions over the features. We show that state-of-the-art performance is achieved in low-resolution data, and our method outperforms previous methods at higher resolution.

Implicit Neural Representations (INR) or neural fields have emerged as a popular framework to encode multimedia signals such as images and radiance fields while retaining high-quality. Recently, learnable feature grids proposed by Instant-NGP have allowed significant speed-up in the training as well as the sampling of INRs by replacing a large neural network with a multi-resolution look-up table of feature vectors and a much smaller neural network. However, these feature grids come at the expense of large memory consumption which can be a bottleneck for storage and streaming applications. In this work, we propose SHACIRA, a simple yet effective task-agnostic framework for compressing such feature grids with no additional post-hoc pruning/quantization stages. We reparameterize feature grids with quantized latent weights and apply entropy regularization in the latent space to achieve high levels of compression across various domains. Quantitative and qualitative results on diverse datasets consisting of images, videos, and radiance fields, show that our approach outperforms existing INR approaches without the need for any large datasets or domain-specific heuristics. Our project page is available at http://shacira.github.io .

A set of vectors \(S \subseteq \mathbb{R}^d\) is \((k_1,\epsilon)\)-clusterable if there are \(k_1\) balls of radius \(\epsilon\) that cover \(S\). A set of vectors \(S \subseteq \mathbb{R}^d\) is \((k_2,\delta)\)-far from being clusterable if there are at least \(k_2\) vectors in \(S\), with all pairwise distances at least \(\delta\). We propose a probabilistic algorithm to distinguish between these two cases. Our algorithm reaches a decision by only looking at the extreme values of a scalar valued hash function, defined by a random field, on \(S\); hence, it is especially suitable in distributed and online settings. An important feature of our method is that the algorithm is oblivious to the number of vectors: in the online setting, for example, the algorithm stores only a constant number of scalars, which is independent of the stream length. We introduce random field hash functions, which are a key ingredient in our paradigm. Random field hash functions generalize locality-sensitive hashing (LSH). In addition to the LSH requirement that nearby vectors are hashed to similar values", our hash function also guarantees that thehash values are (nearly) independent random variables for distant vectors”. We formulate necessary conditions for the kernels which define the random fields applied to our problem, as well as a measure of kernel optimality, for which we provide a bound. Then, we propose a method to construct kernels which approximate the optimal one.

The field of Automatic Machine Learning (AutoML) has recently attained impressive results, including the discovery of state-of-the-art machine learning solutions, such as neural image classifiers. This is often done by applying an evolutionary search method, which samples multiple candidate solutions from a large space and evaluates the quality of each candidate through a long training process. As a result, the search tends to be slow. In this paper, we show that large efficiency gains can be obtained by employing a fast unified functional hash, especially through the functional equivalence caching technique, which we also present. The central idea is to detect by hashing when the search method produces equivalent candidates, which occurs very frequently, and this way avoid their costly re-evaluation. Our hash is “functional” in that it identifies equivalent candidates even if they were represented or coded differently, and it is “unified” in that the same algorithm can hash arbitrary representations; e.g. compute graphs, imperative code, or lambda functions. As evidence, we show dramatic improvements on multiple AutoML domains, including neural architecture search and algorithm discovery. Finally, we consider the effect of hash collisions, evaluation noise, and search distribution through empirical analysis. Altogether, we hope this paper may serve as a guide to hashing techniques in AutoML.

Approximate K nearest neighbor (AKNN) search is a fundamental and challenging problem. We observe that in high-dimensional space, the time consumption of nearly all AKNN algorithms is dominated by that of the distance comparison operations (DCOs). For each operation, it scans full dimensions of an object and thus, runs in linear time wrt the dimensionality. To speed it up, we propose a randomized algorithm named ADSampling which runs in logarithmic time wrt to the dimensionality for the majority of DCOs and succeeds with high probability. In addition, based on ADSampling we develop one general and two algorithm-specific techniques as plugins to enhance existing AKNN algorithms. Both theoretical and empirical studies confirm that: (1) our techniques introduce nearly no accuracy loss and (2) they consistently improve the efficiency.

Conservative Count-Min, an improved version of Count-Min sketch [Cormode, Muthukrishnan 2005], is an online-maintained hashing-based data structure summarizing element frequency information without storing elements themselves. Although several works attempted to analyze the error that can be made by Count-Min, the behavior of this data structure remains poorly understood. In [Fusy, Kucherov 2022], we demonstrated that under the uniform distribution of input elements, the error of conservative Count-Min follows two distinct regimes depending on its load factor. In this work, we provide a series of experimental results providing new insights into the behavior of conservative Count-Min. Our contributions can be seen as twofold. On one hand, we provide a detailed experimental analysis of the behavior of Count-Min sketch in different regimes and under several representative probability distributions of input elements. On the other hand, we demonstrate improvements that can be made by assigning a variable number of hash functions to different elements. This includes, in particular, reduced space of the data structure while still supporting a small error.

In this paper, we analyze hashing from a worst-case perspective. To this end, we study a new property of hash families that is strongly related to d-perfect hashing, namely c-ideality. On the one hand, this notion generalizes the definition of perfect hashing, which has been studied extensively; on the other hand, it provides a direct link to the notion of c-approximativity. We focus on the usually neglected case where the average load \alpha is at least 1 and prove upper and lower parametrized bounds on the minimal size of c-ideal hash families. As an aside, we show how c-ideality helps to analyze the advice complexity of hashing. The concept of advice, introduced a decade ago, lets us measure the information content of an online problem. We prove hashing’s advice complexity to be linear in the hash table size.

A Monotone Minimal Perfect Hash Function (MMPHF) constructed on a set S of keys is a function that maps each key in S to its rank. On keys not in S, the function returns an arbitrary value. Applications range from databases, search engines, data encryption, to pattern-matching algorithms. In this paper, we describe LeMonHash, a new technique for constructing MMPHFs for integers. The core idea of LeMonHash is surprisingly simple and effective: we learn a monotone mapping from keys to their rank via an error-bounded piecewise linear model (the PGM-index), and then we solve the collisions that might arise among keys mapping to the same rank estimate by associating small integers with them in a retrieval data structure (BuRR). On synthetic random datasets, LeMonHash needs 34% less space than the next larger competitor, while achieving about 16 times faster queries. On real-world datasets, the space usage is very close to or much better than the best competitors, while achieving up to 19 times faster queries than the next larger competitor. As far as the construction of LeMonHash is concerned, we get an improvement by a factor of up to 2, compared to the competitor with the next best space usage. We also investigate the case of keys being variable-length strings, introducing the so-called LeMonHash-VL: it needs space within 13% of the best competitors while achieving up to 3 times faster queries than the next larger competitor.

In this work we study Invertible Bloom Lookup Tables (IBLTs) with small failure probabilities. IBLTs are highly versatile data structures that have found applications in set reconciliation protocols, error-correcting codes, and even the design of advanced cryptographic primitives. For storing \(n\) elements and ensuring correctness with probability at least \(1 - \delta\), existing IBLT constructions require \(Ω(n(\frac{log(1/\delta)}{log(n)}+1))\) space and they crucially rely on fully random hash functions. We present new constructions of IBLTs that are simultaneously more space efficient and require less randomness. For storing \(n\) elements with a failure probability of at most \(\delta\), our data structure only requires \(\mathcal{O}(n + log(1/\delta)loglog(1/\delta))\) space and \(\mathcal{O}(log(log(n)/\delta))\)-wise independent hash functions. As a key technical ingredient we show that hashing \(n\) keys with any \(k\)-wise independent hash function \(h:U \to [Cn]\) for some sufficiently large constant \(C\) guarantees with probability \(1 - 2^{-Ω(k)}\) that at least \(n/2\) keys will have a unique hash value. Proving this is highly non-trivial as \(k\) approaches \(n\). We believe that the techniques used to prove this statement may be of independent interest.

The expenses involved in training state-of-the-art deep hashing retrieval models have witnessed an increase due to the adoption of more sophisticated models and large-scale datasets. Dataset Distillation (DD) or Dataset Condensation(DC) focuses on generating smaller synthetic dataset that retains the original information. Nevertheless, existing DD methods face challenges in maintaining a trade-off between accuracy and efficiency. And the state-of-the-art dataset distillation methods can not expand to all deep hashing retrieval methods. In this paper, we propose an efficient condensation framework that addresses these limitations by matching the feature-embedding between synthetic set and real set. Furthermore, we enhance the diversity of features by incorporating the strategies of early-stage augmented models and multi-formation. Extensive experiments provide compelling evidence of the remarkable superiority of our approach, both in terms of performance and efficiency, compared to state-of-the-art baseline methods.

Large-scale embedding-based retrieval (EBR) is the cornerstone of search-related industrial applications. Given a user query, the system of EBR aims to identify relevant information from a large corpus of documents that may be tens or hundreds of billions in size. The storage and computation turn out to be expensive and inefficient with massive documents and high concurrent queries, making it difficult to further scale up. To tackle the challenge, we propose a binary embedding-based retrieval (BEBR) engine equipped with a recurrent binarization algorithm that enables customized bits per dimension. Specifically, we compress the full-precision query and document embeddings, formulated as float vectors in general, into a composition of multiple binary vectors using a lightweight transformation model with residual multilayer perception (MLP) blocks. We can therefore tailor the number of bits for different applications to trade off accuracy loss and cost savings. Importantly, we enable task-agnostic efficient training of the binarization model using a new embedding-to-embedding strategy. We also exploit the compatible training of binary embeddings so that the BEBR engine can support indexing among multiple embedding versions within a unified system. To further realize efficient search, we propose Symmetric Distance Calculation (SDC) to achieve lower response time than Hamming codes. We successfully employed the introduced BEBR to Tencent products, including Sogou, Tencent Video, QQ World, etc. The binarization algorithm can be seamlessly generalized to various tasks with multiple modalities. Extensive experiments on offline benchmarks and online A/B tests demonstrate the efficiency and effectiveness of our method, significantly saving 30%~50% index costs with almost no loss of accuracy at the system level.

A vector database is used to store high-dimensional data that cannot be characterized by traditional DBMS. Although there are not many articles describing existing or introducing new vector database architectures, the approximate nearest neighbor search problem behind vector databases has been studied for a long time, and considerable related algorithmic articles can be found in the literature. This article attempts to comprehensively review relevant algorithms to provide a general understanding of this booming research area. The basis of our framework categorises these studies by the approach of solving ANNS problem, respectively hash-based, tree-based, graph-based and quantization-based approaches. Then we present an overview of existing challenges for vector databases. Lastly, we sketch how vector databases can be combined with large language models and provide new possibilities.

Extreme multilabel classification or XML, is an active area of interest in machine learning. Compared to traditional multilabel classification, here the number of labels is extremely large, hence, the name extreme multilabel classification. Using classical one versus all classification wont scale in this case due to large number of labels, same is true for any other classifiers. Embedding of labels as well as features into smaller label space is an essential first step. Moreover, other issues include existence of head and tail labels, where tail labels are labels which exist in relatively smaller number of given samples. The existence of tail labels creates issues during embedding. This area has invited application of wide range of approaches ranging from bit compression motivated from compressed sensing, tree based embeddings, deep learning based latent space embedding including using attention weights, linear algebra based embeddings such as SVD, clustering, hashing, to name a few. The community has come up with a useful set of metrics to identify correctly the prediction for head or tail labels.

We introduce linear probing hashing schemes that construct a hash table of size \(n\), with constant load factor \(\alpha\), on which the worst-case unsuccessful search time is asymptotically almost surely \(O(log log n)\). The schemes employ two linear probe sequences to find empty cells for the keys. Matching lower bounds on the maximum cluster size produced by any algorithm that uses two linear probe sequences are obtained as well.

Deep hashing has been intensively studied and successfully applied in large-scale image retrieval systems due to its efficiency and effectiveness. Recent studies have recognized that the existence of adversarial examples poses a security threat to deep hashing models, that is, adversarial vulnerability. Notably, it is challenging to efficiently distill reliable semantic representatives for deep hashing to guide adversarial learning, and thereby it hinders the enhancement of adversarial robustness of deep hashing-based retrieval models. Moreover, current researches on adversarial training for deep hashing are hard to be formalized into a unified minimax structure. In this paper, we explore Semantic-Aware Adversarial Training (SAAT) for improving the adversarial robustness of deep hashing models. Specifically, we conceive a discriminative mainstay features learning (DMFL) scheme to construct semantic representatives for guiding adversarial learning in deep hashing. Particularly, our DMFL with the strict theoretical guarantee is adaptively optimized in a discriminative learning manner, where both discriminative and semantic properties are jointly considered. Moreover, adversarial examples are fabricated by maximizing the Hamming distance between the hash codes of adversarial samples and mainstay features, the efficacy of which is validated in the adversarial attack trials. Further, we, for the first time, formulate the formalized adversarial training of deep hashing into a unified minimax optimization under the guidance of the generated mainstay codes. Extensive experiments on benchmark datasets show superb attack performance against the state-of-the-art algorithms, meanwhile, the proposed adversarial training can effectively eliminate adversarial perturbations for trustworthy deep hashing-based retrieval. Our code is available at https://github.com/xandery-geek/SAAT.

In recent years there has been a shift from heuristics-based malware detection towards machine learning, which proves to be more robust in the current heavily adversarial threat landscape. While we acknowledge machine learning to be better equipped to mine for patterns in the increasingly high amounts of similar-looking files, we also note a remarkable scarcity of the data available for similarity-targeted research. Moreover, we observe that the focus in the few related works falls on quantifying similarity in malware, often overlooking the clean data. This one-sided quantification is especially dangerous in the context of detection bypass. We propose to address the deficiencies in the space of similarity research on binary files, starting from EMBER - one of the largest malware classification data sets. We enhance EMBER with similarity information as well as malware class tags, to enable further research in the similarity space. Our contribution is threefold: (1) we publish EMBERSim, an augmented version of EMBER, that includes similarity-informed tags; (2) we enrich EMBERSim with automatically determined malware class tags using the open-source tool AVClass on VirusTotal data and (3) we describe and share the implementation for our class scoring technique and leaf similarity method.

Consistent hashing is used in distributed systems and networking applications to spread data evenly and efficiently across a cluster of nodes. In this paper, we present MementoHash, a novel consistent hashing algorithm that eliminates known limitations of state-of-the-art algorithms while keeping optimal performance and minimal memory usage. We describe the algorithm in detail, provide a pseudo-code implementation, and formally establish its solid theoretical guarantees. To measure the efficacy of MementoHash, we compare its performance, in terms of memory usage and lookup time, to that of state-of-the-art algorithms, namely, AnchorHash, DxHash, and JumpHash. Unlike JumpHash, MementoHash can handle random failures. Moreover, MementoHash does not require fixing the overall capacity of the cluster (as AnchorHash and DxHash do), allowing it to scale indefinitely. The number of removed nodes affects the performance of all the considered algorithms. Therefore, we conduct experiments considering three different scenarios: stable (no removed nodes), one-shot removals (90% of the nodes removed at once), and incremental removals. We report experimental results that averaged a varying number of nodes from ten to one million. Results indicate that our algorithm shows optimal lookup performance and minimal memory usage in its best-case scenario. It behaves better than AnchorHash and DxHash in its average-case scenario and at least as well as those two algorithms in its worst-case scenario. However, the worst-case scenario for MementoHash occurs when more than 70% of the nodes fail, which describes a unlikely scenario. Therefore, MementoHash shows the best performance during the regular life cycle of a cluster.

Representing visual data using compact binary codes is attracting increasing attention as binary codes are used as direct indices into hash table(s) for fast non-exhaustive search. Recent methods show that ranking binary codes using weighted Hamming distance (WHD) rather than Hamming distance (HD) by generating query-adaptive weights for each bit can better retrieve query-related items. However, search using WHD is slower than that using HD. One main challenge is that the complexity of extending a monotone increasing sequence using WHD to probe buckets in hash table(s) for existing methods is at least proportional to the square of the sequence length, while that using HD is proportional to the sequence length. To overcome this challenge, we propose a novel fast non-exhaustive search method using WHD. The key idea is to design a constant sequence extension algorithm to perform each sequence extension in constant computational complexity and the total complexity is proportional to the sequence length, which is justified by theoretical analysis. Experimental results show that our method is faster than other WHD-based search methods. Also, compared with the HD-based non-exhaustive search method, our method has comparable efficiency but retrieves more query-related items for the dataset of up to one billion items.

Embedding-based retrieval methods construct vector indices to search for document representations that are most similar to the query representations. They are widely used in document retrieval due to low latency and decent recall performance. Recent research indicates that deep retrieval solutions offer better model quality, but are hindered by unacceptable serving latency and the inability to support document updates. In this paper, we aim to enhance the vector index with end-to-end deep generative models, leveraging the differentiable advantages of deep retrieval models while maintaining desirable serving efficiency. We propose Model-enhanced Vector Index (MEVI), a differentiable model-enhanced index empowered by a twin-tower representation model. MEVI leverages a Residual Quantization (RQ) codebook to bridge the sequence-to-sequence deep retrieval and embedding-based models. To substantially reduce the inference time, instead of decoding the unique document ids in long sequential steps, we first generate some semantic virtual cluster ids of candidate documents in a small number of steps, and then leverage the well-adapted embedding vectors to further perform a fine-grained search for the relevant documents in the candidate virtual clusters. We empirically show that our model achieves better performance on the commonly used academic benchmarks MSMARCO Passage and Natural Questions, with comparable serving latency to dense retrieval solutions.

In Linear Hashing (\(\mathsf{LH}\)) with \(\beta\) bins on a size \(u\) universe \({\mathcal{U}=\{0,1,\ldots, u-1\}}\), items \(\{x_1,x_2,\ldots, x_n\}\subset \mathcal{U}\) are placed in bins by the hash function $\(x_i\mapsto (ax_i+b)\mod p \mod \beta\)\( for some prime \)p\in [u,2u]\( and randomly chosen integers \)a,b \in [1,p]\(. The “maxload” of \)\mathsf{LH}\( is the number of items assigned to the fullest bin. Expected maxload for a worst-case set of items is a natural measure of how well \)\mathsf{LH}\( distributes items amongst the bins. Fix \)\beta=n\(. Despite \)\mathsf{LH}\(‘s simplicity, bounding \)\mathsf{LH}\(‘s worst-case maxload is extremely challenging. It is well-known that on random inputs \)\mathsf{LH}\( achieves maxload \)Ω\left(\frac{log n}{loglog n}\right)\(; this is currently the best lower bound for \)\mathsf{LH}\(‘s expected maxload. Recently Knudsen established an upper bound of \)\widetilde{O}(n^{1 / 3})\(. The question “Is the worst-case expected maxload of \)\mathsf{LH}\( \)n^{o(1)}$?” is one of the most basic open problems in discrete math. In this paper we propose a set of intermediate open questions to help researchers make progress on this problem. We establish the relationship between these intermediate open questions and make some partial progress on them.

Deep hashing has shown to be a complexity-efficient solution for the Approximate Nearest Neighbor search problem in high dimensional space. Many methods usually build the loss function from pairwise or triplet data points to capture the local similarity structure. Other existing methods construct the similarity graph and consider all points simultaneously. Auto-encoding Twin-bottleneck Hashing is one such method that dynamically builds the graph. Specifically, each input data is encoded into a binary code and a continuous variable, or the so-called twin bottlenecks. The similarity graph is then computed from these binary codes, which get updated consistently during the training. In this work, we generalize the original model into a supervised deep hashing network by incorporating the label information. In addition, we examine the differences of codes structure between these two networks and consider the class imbalance problem especially in multi-labeled datasets. Experiments on three datasets yield statistically significant improvement against the original model. Results are also comparable and competitive to other supervised methods.

Searching on bipartite graphs is basal and versatile to many real-world Web applications, e.g., online recommendation, database retrieval, and query-document searching. Given a query node, the conventional approaches rely on the similarity matching with the vectorized node embeddings in the continuous Euclidean space. To efficiently manage intensive similarity computation, developing hashing techniques for graph structured data has recently become an emerging research direction. Despite the retrieval efficiency in Hamming space, prior work is however confronted with catastrophic performance decay. In this work, we investigate the problem of hashing with Graph Convolutional Network on bipartite graphs for effective Top-N search. We propose an end-to-end Bipartite Graph Convolutional Hashing approach, namely BGCH, which consists of three novel and effective modules: (1) adaptive graph convolutional hashing, (2) latent feature dispersion, and (3) Fourier serialized gradient estimation. Specifically, the former two modules achieve the substantial retention of the structural information against the inevitable information loss in hash encoding; the last module develops Fourier Series decomposition to the hashing function in the frequency domain mainly for more accurate gradient estimation. The extensive experiments on six real-world datasets not only show the performance superiority over the competing hashing-based counterparts, but also demonstrate the effectiveness of all proposed model components contained therein.

For avoiding the exposure of plaintexts in cloud environments, some homomorphic hashing algorithms have been proposed to generate the hash value of each plaintext, and cloud environments only store the hash values and calculate the hash values for future needs. However, longer hash value generation time and longer hash value summary time may be required by these homomorphic hashing algorithms with higher security strengths. Therefore, this study proposes a homomorphic hashing based on elliptic curve cryptography (ECC) to provide a homomorphic hashing function in accordance with the characteristics of ECC. Furthermore, mathematical models and practical cases have been given to prove the proposed method. In experiments, the results show that the proposed method have higher efficiency with different security strengths.

We present a video decomposition method that facilitates layer-based editing of videos with spatiotemporally varying lighting and motion effects. Our neural model decomposes an input video into multiple layered representations, each comprising a 2D texture map, a mask for the original video, and a multiplicative residual characterizing the spatiotemporal variations in lighting conditions. A single edit on the texture maps can be propagated to the corresponding locations in the entire video frames while preserving other contents’ consistencies. Our method efficiently learns the layer-based neural representations of a 1080p video in 25s per frame via coordinate hashing and allows real-time rendering of the edited result at 71 fps on a single GPU. Qualitatively, we run our method on various videos to show its effectiveness in generating high-quality editing effects. Quantitatively, we propose to adopt feature-tracking evaluation metrics for objectively assessing the consistency of video editing. Project page: https://lightbulb12294.github.io/hashing-nvd/

Given a data stream \(\mathcal{A} = \langle a_1, a_2, \ldots, a_m \rangle\) of \(m\) elements where each \(a_i \in [n]\), the Distinct Elements problem is to estimate the number of distinct elements in \(\mathcal{A}\).Distinct Elements has been a subject of theoretical and empirical investigations over the past four decades resulting in space optimal algorithms for it.All the current state-of-the-art algorithms are, however, beyond the reach of an undergraduate textbook owing to their reliance on the usage of notions such as pairwise independence and universal hash functions. We present a simple, intuitive, sampling-based space-efficient algorithm whose description and the proof are accessible to undergraduates with the knowledge of basic probability theory.

Cuckoo hashing is a powerful primitive that enables storing items using small space with efficient querying. At a high level, cuckoo hashing maps \(n\) items into \(b\) entries storing at most \(\ell\) items such that each item is placed into one of \(k\) randomly chosen entries. Additionally, there is an overflow stash that can store at most \(s\) items. Many cryptographic primitives rely upon cuckoo hashing to privately embed and query data where it is integral to ensure small failure probability when constructing cuckoo hashing tables as it directly relates to the privacy guarantees. As our main result, we present a more query-efficient cuckoo hashing construction using more hash functions. For construction failure probability \(\epsilon\), the query overhead of our scheme is \(O(1 + \sqrt{log(1/\epsilon)/log n})\). Our scheme has quadratically smaller query overhead than prior works for any target failure probability \(\epsilon\). We also prove lower bounds matching our construction. Our improvements come from a new understanding of the locality of cuckoo hashing failures for small sets of items. We also initiate the study of robust cuckoo hashing where the input set may be chosen with knowledge of the hash functions. We present a cuckoo hashing scheme using more hash functions with query overhead \(\tilde{O}(log \lambda)\) that is robust against poly\((\lambda)\) adversaries. Furthermore, we present lower bounds showing that this construction is tight and that extending previous approaches of large stashes or entries cannot obtain robustness except with \(Ω(n)\) query overhead. As applications of our results, we obtain improved constructions for batch codes and PIR. In particular, we present the most efficient explicit batch code and blackbox reduction from single-query PIR to batch PIR.

Since 2016, NIST has been assessing lightweight encryption methods, and, in 2022, NIST published the final 10: ASCON, Elephant, GIFT-COFB, Grain128-AEAD, ISAP, Photon-Beetle, Romulus, Sparkle, TinyJambu, and Xoodyak. At the time that the article was written, NISC announced ASCOn as the chosen method that will be published as NIST’S lightweight cryptography standard later in 2023. In this article, we provide a comparison between these methods in terms of energy efficiency, time for encryption, and time for hashing.

We consider the problem of counting 4-cycles (\(C_4\)) in an undirected graph \(G\) of \(n\) vertices and \(m\) edges (in bipartite graphs, 4-cycles are also often referred to as \(\textit{butterflies}\)). Most recently, Wang et al. (2019, 2022) developed algorithms for this problem based on hash tables and sorting the graph by degree. Their algorithm takes \(O(m\bar\delta)\) expected time and \(O(m)\) space, where \(\bar \delta \leq O(\sqrt{m})\) is the \(\textit{average degeneracy}\) parameter introduced by Burkhardt, Faber \& Harris (2020). We develop a streamlined version of this algorithm requiring \(O(m\bar\delta)\) time and precisely \(n\) words of space. It has several practical improvements and optimizations; for example, it is fully deterministic, does not require any auxiliary storage or sorting of the input graph, and uses only addition and array access in its inner loops. Our algorithm is very simple and easily adapted to count 4-cycles incident to each vertex and edge. Empirical tests demonstrate that our array-based approach is \(4\times\) – \(7\times\) faster on average compared to popular hash table implementations.

We present a new approach for computing compact sketches that can be used to approximate the inner product between pairs of high-dimensional vectors. Based on the Weighted MinHash algorithm, our approach admits strong accuracy guarantees that improve on the guarantees of popular linear sketching approaches for inner product estimation, such as CountSketch and Johnson-Lindenstrauss projection. Specifically, while our method admits guarantees that exactly match linear sketching for dense vectors, it yields significantly lower error for sparse vectors with limited overlap between non-zero entries. Such vectors arise in many applications involving sparse data. They are also important in increasingly popular dataset search applications, where inner product sketches are used to estimate data covariance, conditional means, and other quantities involving columns in unjoined tables. We complement our theoretical results by showing that our approach empirically outperforms existing linear sketches and unweighted hashing-based sketches for sparse vectors.

Detecting visually similar images is a particularly useful attribute to look to when calculating product recommendations. Embedding similarity, which utilizes pre-trained computer vision models to extract high-level image features, has demonstrated remarkable efficacy in identifying images with similar compositions. However, there is a lack of methods for evaluating the embeddings generated by these models, as conventional loss and performance metrics do not adequately capture their performance in image similarity search tasks. In this paper, we evaluate the viability of the image embeddings from numerous pre-trained computer vision models using a novel approach named CorrEmbed. Our approach computes the correlation between distances in image embeddings and distances in human-generated tag vectors. We extensively evaluate numerous pre-trained Torchvision models using this metric, revealing an intuitive relationship of linear scaling between ImageNet1k accuracy scores and tag-correlation scores. Importantly, our method also identifies deviations from this pattern, providing insights into how different models capture high-level image features. By offering a robust performance evaluation of these pre-trained models, CorrEmbed serves as a valuable tool for researchers and practitioners seeking to develop effective, data-driven approaches to similar item recommendations in fashion retail.

Hashing is a common technique used in data processing, with a strong impact on the time and resources spent on computation. Hashing also affects the applicability of theoretical results that often assume access to (unrealistic) uniform/fully-random hash functions. In this paper, we are concerned with designing hash functions that are practical and come with strong theoretical guarantees on their performance. To this end, we present tornado tabulation hashing, which is simple, fast, and exhibits a certain full, local randomness property that provably makes diverse algorithms perform almost as if (abstract) fully-random hashing was used. For example, this includes classic linear probing, the widely used HyperLogLog algorithm of Flajolet, Fusy, Gandouet, Meunier [AOFA 97] for counting distinct elements, and the one-permutation hashing of Li, Owen, and Zhang [NIPS 12] for large-scale machine learning. We also provide a very efficient solution for the classical problem of obtaining fully-random hashing on a fixed (but unknown to the hash function) set of \(n\) keys using \(O(n)\) space. As a consequence, we get more efficient implementations of the splitting trick of Dietzfelbinger and Rink [ICALP’09] and the succinct space uniform hashing of Pagh and Pagh [SICOMP’08]. Tornado tabulation hashing is based on a simple method to systematically break dependencies in tabulation-based hashing techniques.

The statistical distribution of content uploaded and searched on media sharing sites changes over time due to seasonal, sociological and technical factors. We investigate the impact of this “content drift” for large-scale similarity search tools, based on nearest neighbor search in embedding space. Unless a costly index reconstruction is performed frequently, content drift degrades the search accuracy and efficiency. The degradation is especially severe since, in general, both the query and database distributions change. We introduce and analyze real-world image and video datasets for which temporal information is available over a long time period. Based on the learnings, we devise DeDrift, a method that updates embedding quantizers to continuously adapt large-scale indexing structures on-the-fly. DeDrift almost eliminates the accuracy degradation due to the query and database content drift while being up to 100x faster than a full index reconstruction.

Large-scale visual localization systems continue to rely on 3D point clouds built from image collections using structure-from-motion. While the 3D points in these models are represented using local image features, directly matching a query image’s local features against the point cloud is challenging due to the scale of the nearest-neighbor search problem. Many recent approaches to visual localization have thus proposed a hybrid method, where first a global (per image) embedding is used to retrieve a small subset of database images, and local features of the query are matched only against those. It seems to have become common belief that global embeddings are critical for said image-retrieval in visual localization, despite the significant downside of having to compute two feature types for each query image. In this paper, we take a step back from this assumption and propose Constrained Approximate Nearest Neighbors (CANN), a joint solution of k-nearest-neighbors across both the geometry and appearance space using only local features. We first derive the theoretical foundation for k-nearest-neighbor retrieval across multiple metrics and then showcase how CANN improves visual localization. Our experiments on public localization benchmarks demonstrate that our method significantly outperforms both state-of-the-art global feature-based retrieval and approaches using local feature aggregation schemes. Moreover, it is an order of magnitude faster in both index and query time than feature aggregation schemes for these datasets. Code: \url{https://github.com/google-research/google-research/tree/master/cann}

Nowadays, data is represented by vectors. Retrieving those vectors, among millions and billions, that are similar to a given query is a ubiquitous problem, known as similarity search, of relevance for a wide range of applications. Graph-based indices are currently the best performing techniques for billion-scale similarity search. However, their random-access memory pattern presents challenges to realize their full potential. In this work, we present new techniques and systems for creating faster and smaller graph-based indices. To this end, we introduce a novel vector compression method, Locally-adaptive Vector Quantization (LVQ), that uses per-vector scaling and scalar quantization to improve search performance with fast similarity computations and a reduced effective bandwidth, while decreasing memory footprint and barely impacting accuracy. LVQ, when combined with a new high-performance computing system for graph-based similarity search, establishes the new state of the art in terms of performance and memory footprint. For billions of vectors, LVQ outcompetes the second-best alternatives: (1) in the low-memory regime, by up to 20.7x in throughput with up to a 3x memory footprint reduction, and (2) in the high-throughput regime by 5.8x with 1.4x less memory.

Given a set of points labeled with \(k\) labels, we introduce the heat map sorting problem as reordering and merging the points and dimensions while preserving the clusters (labels). A cluster is preserved if it remains connected, i.e., if it is not split into several clusters and no two clusters are merged. We prove the problem is NP-hard and we give a fixed-parameter algorithm with a constant number of rounds in the massively parallel computation model, where each machine has a sublinear memory and the total memory of the machines is linear. We give an approximation algorithm for a NP-hard special case of the problem. We empirically compare our algorithm with k-means and density-based clustering (DBSCAN) using a dimensionality reduction via locality-sensitive hashing on several directed and undirected graphs of email and computer networks.

40 years ago, Conway and Sloane proposed using the highly symmetrical Coxeter-Todd lattice \(K_{12}\) for quantization, and estimated its second moment. Since then, all published lists identify \(K_{12}\) as the best 12-dimensional lattice quantizer. Surprisingly, \(K_{12}\) is not optimal: we construct two new 12-dimensional lattices with lower normalized second moments. The new lattices are obtained by gluing together 6-dimensional lattices.

Recent advancements in deep learning and computer vision have led to widespread use of deep neural networks to extract building footprints from remote-sensing imagery. The success of such methods relies on the availability of large databases of high-resolution remote sensing images with high-quality annotations. The CrowdAI Mapping Challenge Dataset is one of these datasets that has been used extensively in recent years to train deep neural networks. This dataset consists of \( \sim\ \)280k training images and \( \sim\ \)60k testing images, with polygonal building annotations for all images. However, issues such as low-quality and incorrect annotations, extensive duplication of image samples, and data leakage significantly reduce the utility of deep neural networks trained on the dataset. Therefore, it is an imperative pre-condition to adopt a data validation pipeline that evaluates the quality of the dataset prior to its use. To this end, we propose a drop-in pipeline that employs perceptual hashing techniques for efficient de-duplication of the dataset and identification of instances of data leakage between training and testing splits. In our experiments, we demonstrate that nearly 250k(\( \sim\ \)90%) images in the training split were identical. Moreover, our analysis on the validation split demonstrates that roughly 56k of the 60k images also appear in the training split, resulting in a data leakage of 93%. The source code used for the analysis and de-duplication of the CrowdAI Mapping Challenge dataset is publicly available at https://github.com/yeshwanth95/CrowdAI_Hash_and_search .

Locality-sensitive hashing (LSH) is an effective randomized technique widely used in many machine learning tasks. The cost of hashing is proportional to data dimensions, and thus often the performance bottleneck when dimensionality is high and the number of hash functions involved is large. Surprisingly, however, little work has been done to improve the efficiency of LSH computation. In this paper, we design a simple yet efficient LSH scheme, named FastLSH, under l2 norm. By combining random sampling and random projection, FastLSH reduces the time complexity from O(n) to O(m) (m<n), where n is the data dimensionality and m is the number of sampled dimensions. Moreover, FastLSH has provable LSH property, which distinguishes it from the non-LSH fast sketches. We conduct comprehensive experiments over a collection of real and synthetic datasets for the nearest neighbor search task. Experimental results demonstrate that FastLSH is on par with the state-of-the-arts in terms of answer quality, space occupation and query efficiency, while enjoying up to 80x speedup in hash function evaluation. We believe that FastLSH is a promising alternative to the classic LSH scheme.

Approximate nearest neighbour (ANN) search is an essential component of search engines, recommendation systems, etc. Many recent works focus on learning-based data-distribution-dependent hashing and achieve good retrieval performance. However, due to increasing demand for users’ privacy and security, we often need to remove users’ data information from Machine Learning (ML) models to satisfy specific privacy and security requirements. This need requires the ANN search algorithm to support fast online data deletion and insertion. Current learning-based hashing methods need retraining the hash function, which is prohibitable due to the vast time-cost of large-scale data. To address this problem, we propose a novel data-dependent hashing method named unfolded self-reconstruction locality-sensitive hashing (USR-LSH). Our USR-LSH unfolded the optimization update for instance-wise data reconstruction, which is better for preserving data information than data-independent LSH. Moreover, our USR-LSH supports fast online data deletion and insertion without retraining. To the best of our knowledge, we are the first to address the machine unlearning of retrieval problems. Empirically, we demonstrate that USR-LSH outperforms the state-of-the-art data-distribution-independent LSH in ANN tasks in terms of precision and recall. We also show that USR-LSH has significantly faster data deletion and insertion time than learning-based data-dependent hashing.

This paper demonstrates the utility of organized numerical representations of genes in research involving flat string gene formats (i.e., FASTA/FASTQ5). FASTA/FASTQ files have several current limitations, such as their large file sizes, slow processing speeds for mapping and alignment, and contextual dependencies. These challenges significantly hinder investigations and tasks that involve finding similar sequences. The solution lies in transforming sequences into an alternative representation that facilitates easier clustering into similar groups compared to the raw sequences themselves. By assigning a unique vector embedding to each short sequence, it is possible to more efficiently cluster and improve upon compression performance for the string representations of cDNA libraries. Furthermore, through learning alternative coordinate vector embeddings based on the contexts of codon triplets, we can demonstrate clustering based on amino acid properties. Finally, using this sequence embedding method to encode barcodes and cDNA sequences, we can improve the time complexity of the similarity search by coupling vector embeddings with an algorithm that determines the proximity of vectors in Euclidean space; this allows us to perform sequence similarity searches in a quicker and more modular fashion.

Recently proposed systems aim at achieving privacy using locality-sensitive hashing. We show how these approaches fail by presenting attacks against two such systems: Google’s FLoC proposal for privacy-preserving targeted advertising and the MinHash Hierarchy, a system for processing mobile users’ traffic behavior in a privacy-preserving way. Our attacks refute the pre-image resistance, anonymity, and privacy guarantees claimed for these systems. In the case of FLoC, we show how to deanonymize users using Sybil attacks and to reconstruct 10% or more of the browsing history for 30% of its users using Generative Adversarial Networks. We achieve this only analyzing the hashes used by FLoC. For MinHash, we precisely identify the movement of a subset of individuals and, on average, we can limit users’ movement to just 10% of the possible geographic area, again using just the hashes. In addition, we refute their differential privacy claims.

We study efficient mechanisms for differentially private kernel density estimation (DP-KDE). Prior work for the Gaussian kernel described algorithms that run in time exponential in the number of dimensions \(d\). This paper breaks the exponential barrier, and shows how the KDE can privately be approximated in time linear in \(d\), making it feasible for high-dimensional data. We also present improved bounds for low-dimensional data. Our results are obtained through a general framework, which we term Locality Sensitive Quantization (LSQ), for constructing private KDE mechanisms where existing KDE approximation techniques can be applied. It lets us leverage several efficient non-private KDE methods – like Random Fourier Features, the Fast Gauss Transform, and Locality Sensitive Hashing – and ``privatize’’ them in a black-box manner. Our experiments demonstrate that our resulting DP-KDE mechanisms are fast and accurate on large datasets in both high and low dimensions.

A Perfect Hash Function (PHF) is a hash function that has no collisions on a given input set. PHFs can be used for space efficient storage of data in an array, or for determining a compact representative of each object in the set. In this paper, we present the PHF construction algorithm SicHash - Small Irregular Cuckoo Tables for Perfect Hashing. At its core, SicHash uses a known technique: It places objects in a cuckoo hash table and then stores the final hash function choice of each object in a retrieval data structure. We combine the idea with irregular cuckoo hashing, where each object has a different number of hash functions. Additionally, we use many small tables that we overload beyond their asymptotic maximum load factor. The most space efficient competitors often use brute force methods to determine the PHFs. SicHash provides a more direct construction algorithm that only rarely needs to recompute parts. Our implementation improves the state of the art in terms of space usage versus construction time for a wide range of configurations. At the same time, it provides very fast queries.

Sets have been used for modeling various types of objects (e.g., a document as the set of keywords in it and a customer as the set of the items that she has purchased). Measuring similarity (e.g., Jaccard Index) between sets has been a key building block of a wide range of applications, including, plagiarism detection, recommendation, and graph compression. However, as sets have grown in numbers and sizes, the computational cost and storage required for set similarity computation have become substantial, and this has led to the development of hashing and sketching based solutions. In this work, we propose Set2Box, a learning-based approach for compressed representations of sets from which various similarity measures can be estimated accurately in constant time. The key idea is to represent sets as boxes to precisely capture overlaps of sets. Additionally, based on the proposed box quantization scheme, we design Set2Box+, which yields more concise but more accurate box representations of sets. Through extensive experiments on 8 real-world datasets, we show that, compared to baseline approaches, Set2Box+ is (a) Accurate: achieving up to 40.8X smaller estimation error while requiring 60% fewer bits to encode sets, (b) Concise: yielding up to 96.8X more concise representations with similar estimation error, and (c) Versatile: enabling the estimation of four set-similarity measures from a single representation of each set.

We propose a Composite Code Sparse Autoencoder (CCSA) approach for Approximate Nearest Neighbor (ANN) search of document representations based on Siamese-BERT models. In Information Retrieval (IR), the ranking pipeline is generally decomposed in two stages: the first stage focus on retrieving a candidate set from the whole collection. The second stage re-ranks the candidate set by relying on more complex models. Recently, Siamese-BERT models have been used as first stage ranker to replace or complement the traditional bag-of-word models. However, indexing and searching a large document collection require efficient similarity search on dense vectors and this is why ANN techniques come into play. Since composite codes are naturally sparse, we first show how CCSA can learn efficient parallel inverted index thanks to an uniformity regularizer. Second, CCSA can be used as a binary quantization method and we propose to combine it with the recent graph based ANN techniques. Our experiments on MSMARCO dataset reveal that CCSA outperforms IVF with product quantization. Furthermore, CCSA binary quantization is beneficial for the index size, and memory usage for the graph-based HNSW method, while maintaining a good level of recall and MRR. Third, we compare with recent supervised quantization methods for image retrieval and find that CCSA is able to outperform them.

This paper considers the basic question of how strong of a probabilistic guarantee can a hash table, storing \(n\) \((1 + \Theta(1)) log n\)-bit key/value pairs, offer? Past work on this question has been bottlenecked by limitations of the known families of hash functions: The only hash tables to achieve failure probabilities less than \(1 / 2^{\polylog n}\) require access to fully-random hash functions – if the same hash tables are implemented using the known explicit families of hash functions, their failure probabilities become \(1 / \poly(n)\). To get around these obstacles, we show how to construct a randomized data structure that has the same guarantees as a hash table, but that avoids the direct use of hash functions. Building on this, we are able to construct a hash table using \(O(n)\) random bits that achieves failure probability \(1 / n^{n^{1 - \epsilon}}\) for an arbitrary positive constant \(\epsilon\). In fact, we show that this guarantee can even be achieved by a succinct dictionary, that is, by a dictionary that uses space within a \(1 + o(1)\) factor of the information-theoretic optimum. Finally we also construct a succinct hash table whose probabilistic guarantees fall on a different extreme, offering a failure probability of \(1 / \poly(n)\) while using only \(\tilde{O}(log n)\) random bits. This latter result matches (up to low-order terms) a guarantee previously achieved by Dietzfelbinger et al., but with increased space efficiency and with several surprising technical components.

We introduce PaCHash, a hash table that stores its objects contiguously in an array without intervening space, even if the objects have variable size. In particular, each object can be compressed using standard compression techniques. A small search data structure allows locating the objects in constant expected time. PaCHash is most naturally described as a static external hash table where it needs a constant number of bits of internal memory per block of external memory. Here, in some sense, PaCHash beats a lower bound on the space consumption of k-perfect hashing. An implementation for fast SSDs needs about 5 bits of internal memory per block of external memory, requires only one disk access (of variable length) per search operation, and has small internal search overhead compared to the disk access cost. Our experiments show that it has lower space consumption than all previous approaches even when considering objects of identical size.

Kterm Hashing provides an innovative approach to novelty detection on massive data streams. Previous research focused on maximizing the efficiency of Kterm Hashing and succeeded in scaling First Story Detection to Twitter-size data stream without sacrificing detection accuracy. In this paper, we focus on improving the effectiveness of Kterm Hashing. Traditionally, all kterms are considered as equally important when calculating a document’s degree of novelty with respect to the past. We believe that certain kterms are more important than others and hypothesize that uniform kterm weights are sub-optimal for determining novelty in data streams. To validate our hypothesis, we parameterize Kterm Hashing by assigning weights to kterms based on their characteristics. Our experiments apply Kterm Hashing in a First Story Detection setting and reveal that parameterized Kterm Hashing can surpass state-of-the-art detection accuracy and significantly outperform the uniformly weighted approach.

Data similarity (or distance) computation is a fundamental research topic which fosters a variety of similarity-based machine learning and data mining applications. In big data analytics, it is impractical to compute the exact similarity of data instances due to high computational cost. To this end, the Locality Sensitive Hashing (LSH) technique has been proposed to provide accurate estimators for various similarity measures between sets or vectors in an efficient manner without the learning process. Structured data (e.g., sequences, trees and graphs), which are composed of elements and relations between the elements, are commonly seen in the real world, but the traditional LSH algorithms cannot preserve the structure information represented as relations between elements. In order to conquer the issue, researchers have been devoted to the family of the hierarchical LSH algorithms. In this paper, we explore the present progress of the research into hierarchical LSH from the following perspectives: 1) Data structures, where we review various hierarchical LSH algorithms for three typical data structures and uncover their inherent connections; 2) Applications, where we review the hierarchical LSH algorithms in multiple application scenarios; 3) Challenges, where we discuss some potential challenges as future directions.

The in-memory approximate nearest neighbor search (ANNS) algorithms have achieved great success for fast high-recall query processing, but are extremely inefficient when handling hybrid queries with unstructured (i.e., feature vectors) and structured (i.e., related attributes) constraints. In this paper, we present HQANN, a simple yet highly efficient hybrid query processing framework which can be easily embedded into existing proximity graph-based ANNS algorithms. We guarantee both low latency and high recall by leveraging navigation sense among attributes and fusing vector similarity search with attribute filtering. Experimental results on both public and in-house datasets demonstrate that HQANN is 10x faster than the state-of-the-art hybrid ANNS solutions to reach the same recall quality and its performance is hardly affected by the complexity of attributes. It can reach 99\% recall@10 in just around 50 microseconds On GLOVE-1.2M with thousands of attribute constraints.

As a crucial approach for compact representation learning, hashing has achieved great success in effectiveness and efficiency. Numerous heuristic Hamming space metric learning objectives are designed to obtain high-quality hash codes. Nevertheless, a theoretical analysis of criteria for learning good hash codes remains largely unexploited. In this paper, we prove that inter-class distinctiveness and intra-class compactness among hash codes determine the lower bound of hash codes’ performance. Promoting these two characteristics could lift the bound and improve hash learning. We then propose a surrogate model to fully exploit the above objective by estimating the posterior of hash codes and controlling it, which results in a low-bias optimization. Extensive experiments reveal the effectiveness of the proposed method. By testing on a series of hash-models, we obtain performance improvements among all of them, with an up to \(26.5\%\) increase in mean Average Precision and an up to \(20.5\%\) increase in accuracy. Our code is publicly available at https://github.com/VL-Group/LBHash.

Unsupervised image retrieval aims to learn an efficient retrieval system without expensive data annotations, but most existing methods rely heavily on handcrafted feature descriptors or pre-trained feature extractors. To minimize human supervision, recent advance proposes deep fully unsupervised image retrieval aiming at training a deep model from scratch to jointly optimize visual features and quantization codes. However, existing approach mainly focuses on instance contrastive learning without considering underlying semantic structure information, resulting in sub-optimal performance. In this work, we propose a novel self-supervised consistent quantization approach to deep fully unsupervised image retrieval, which consists of part consistent quantization and global consistent quantization. In part consistent quantization, we devise part neighbor semantic consistency learning with codeword diversity regularization. This allows to discover underlying neighbor structure information of sub-quantized representations as self-supervision. In global consistent quantization, we employ contrastive learning for both embedding and quantized representations and fuses these representations for consistent contrastive regularization between instances. This can make up for the loss of useful representation information during quantization and regularize consistency between instances. With a unified learning objective of part and global consistent quantization, our approach exploits richer self-supervision cues to facilitate model learning. Extensive experiments on three benchmark datasets show the superiority of our approach over the state-of-the-art methods.

Product quantization (PQ) coupled with a space rotation, is widely used in modern approximate nearest neighbor (ANN) search systems to significantly compress the disk storage for embeddings and speed up the inner product computation. Existing rotation learning methods, however, minimize quantization distortion for fixed embeddings, which are not applicable to an end-to-end training scenario where embeddings are updated constantly. In this paper, based on geometric intuitions from Lie group theory, in particular the special orthogonal group \(SO(n)\), we propose a family of block Givens coordinate descent algorithms to learn rotation matrix that are provably convergent on any convex objectives. Compared to the state-of-the-art SVD method, the Givens algorithms are much more parallelizable, reducing runtime by orders of magnitude on modern GPUs, and converge more stably according to experimental studies. They further improve upon vanilla product quantization significantly in an end-to-end training scenario.

Hashing has been widely researched to solve the large-scale approximate nearest neighbor search problem owing to its time and storage superiority. In recent years, a number of online hashing methods have emerged, which can update the hash functions to adapt to the new stream data and realize dynamic retrieval. However, existing online hashing methods are required to update the whole database with the latest hash functions when a query arrives, which leads to low retrieval efficiency with the continuous increase of the stream data. On the other hand, these methods ignore the supervision relationship among the examples, especially in the multi-label case. In this paper, we propose a novel Fast Online Hashing (FOH) method which only updates the binary codes of a small part of the database. To be specific, we first build a query pool in which the nearest neighbors of each central point are recorded. When a new query arrives, only the binary codes of the corresponding potential neighbors are updated. In addition, we create a similarity matrix which takes the multi-label supervision information into account and bring in the multi-label projection loss to further preserve the similarity among the multi-label data. The experimental results on two common benchmarks show that the proposed FOH can achieve dramatic superiority on query time up to 6.28 seconds less than state-of-the-art baselines with competitive retrieval accuracy.

State-of-the-art algorithms for Approximate Nearest Neighbor Search (ANNS) such as DiskANN, FAISS-IVF, and HNSW build data dependent indices that offer substantially better accuracy and search efficiency over data-agnostic indices by overfitting to the index data distribution. When the query data is drawn from a different distribution - e.g., when index represents image embeddings and query represents textual embeddings - such algorithms lose much of this performance advantage. On a variety of datasets, for a fixed recall target, latency is worse by an order of magnitude or more for Out-Of-Distribution (OOD) queries as compared to In-Distribution (ID) queries. The question we address in this work is whether ANNS algorithms can be made efficient for OOD queries if the index construction is given access to a small sample set of these queries. We answer positively by presenting OOD-DiskANN, which uses a sparing sample (1% of index set size) of OOD queries, and provides up to 40% improvement in mean query latency over SoTA algorithms of a similar memory footprint. OOD-DiskANN is scalable and has the efficiency of graph-based ANNS indices. Some of our contributions can improve query efficiency for ID queries as well.

Image retrieval has become an increasingly appealing technique with broad multimedia application prospects, where deep hashing serves as the dominant branch towards low storage and efficient retrieval. In this paper, we carried out in-depth investigations on metric learning in deep hashing for establishing a powerful metric space in multi-label scenarios, where the pair loss suffers high computational overhead and converge difficulty, while the proxy loss is theoretically incapable of expressing the profound label dependencies and exhibits conflicts in the constructed hypersphere space. To address the problems, we propose a novel metric learning framework with Hybrid Proxy-Pair Loss (HyP{{ ‘{{’ }}^2{{ ‘}}’ }} Loss) that constructs an expressive metric space with efficient training complexity w.r.t. the whole dataset. The proposed HyP{{ ‘{{’ }}^2{{ ‘}}’ }} Loss focuses on optimizing the hypersphere space by learnable proxies and excavating data-to-data correlations of irrelevant pairs, which integrates sufficient data correspondence of pair-based methods and high-efficiency of proxy-based methods. Extensive experiments on four standard multi-label benchmarks justify the proposed method outperforms the state-of-the-art, is robust among different hash bits and achieves significant performance gains with a faster, more stable convergence speed. Our code is available at https://github.com/JerryXu0129/HyP2-Loss.

Finding similar data in high-dimensional spaces is one of the important tasks in multimedia applications. Approaches introduced to find exact searching techniques often use tree-based index structures which are known to suffer from the curse of the dimensionality problem that limits their performance. Approximate searching techniques prefer performance over accuracy and they return good enough results while achieving a better performance. Locality Sensitive Hashing (LSH) is one of the most popular approximate nearest neighbor search techniques for high-dimensional spaces. One of the most time-consuming processes in LSH is to find the neighboring points in the projected spaces. An improved LSH-based index structure, called radius-optimized Locality Sensitive Hashing (roLSH) has been proposed to utilize Machine Learning and efficiently find these neighboring points; thus, further improve the overall performance of LSH. In this paper, we extend roLSH by experimentally studying the effect of different types of famous Machine Learning techniques on overall performance. We compare ten regression techniques on four real-world datasets and show that Neural Network-based techniques are the best fit to be used in roLSH as their accuracy and performance trade-off are the best compared to the other techniques.

Both supervised and unsupervised machine learning algorithms have been used to learn partition-based index structures for approximate nearest neighbor (ANN) search. Existing supervised algorithms formulate the learning task as finding a partition in which the nearest neighbors of a training set point belong to the same partition element as the point itself, so that the nearest neighbor candidates can be retrieved by naive lookup or backtracking search. We formulate candidate set selection in ANN search directly as a multilabel classification problem where the labels correspond to the nearest neighbors of the query point, and interpret the partitions as partitioning classifiers for solving this task. Empirical results suggest that the natural classifier based on this interpretation leads to strictly improved performance when combined with any unsupervised or supervised partitioning strategy. We also prove a sufficient condition for consistency of a partitioning classifier for ANN search, and illustrate the result by verifying this condition for chronological \(k\)-d trees.

Image retrieval has become an increasingly appealing technique with broad multimedia application prospects, where deep hashing serves as the dominant branch towards low storage and efficient retrieval. In this paper, we carried out in-depth investigations on metric learning in deep hashing for establishing a powerful metric space in multi-label scenarios, where the pair loss suffers high computational overhead and converge difficulty, while the proxy loss is theoretically incapable of expressing the profound label dependencies and exhibits conflicts in the constructed hypersphere space. To address the problems, we propose a novel metric learning framework with Hybrid Proxy-Pair Loss (HyP\(^2\) Loss) that constructs an expressive metric space with efficient training complexity w.r.t. the whole dataset. The proposed HyP\(^2\) Loss focuses on optimizing the hypersphere space by learnable proxies and excavating data-to-data correlations of irrelevant pairs, which integrates sufficient data correspondence of pair-based methods and high-efficiency of proxy-based methods. Extensive experiments on four standard multi-label benchmarks justify the proposed method outperforms the state-of-the-art, is robust among different hash bits and achieves significant performance gains with a faster, more stable convergence speed. Our code is available at https://github.com/JerryXu0129/HyP2-Loss.

This paper investigates a new yet challenging problem called Reverse \(k\)-Maximum Inner Product Search (R\(k\)MIPS). Given a query (item) vector, a set of item vectors, and a set of user vectors, the problem of R\(k\)MIPS aims to find a set of user vectors whose inner products with the query vector are one of the \(k\) largest among the query and item vectors. We propose the first subquadratic-time algorithm, i.e., Shifting-aware Asymmetric Hashing (SAH), to tackle the R\(k\)MIPS problem. To speed up the Maximum Inner Product Search (MIPS) on item vectors, we design a shifting-invariant asymmetric transformation and develop a novel sublinear-time Shifting-Aware Asymmetric Locality Sensitive Hashing (SA-ALSH) scheme. Furthermore, we devise a new blocking strategy based on the Cone-Tree to effectively prune user vectors (in a batch). We prove that SAH achieves a theoretical guarantee for solving the RMIPS problem. Experimental results on five real-world datasets show that SAH runs 4\(\sim\)8\(\times\) faster than the state-of-the-art methods for R\(k\)MIPS while achieving F1-scores of over 90\%. The code is available at \url{https://github.com/HuangQiang/SAH}.

Due to its powerful feature learning capability and high efficiency, deep hashing has achieved great success in large-scale image retrieval. Meanwhile, extensive works have demonstrated that deep neural networks (DNNs) are susceptible to adversarial examples, and exploring adversarial attack against deep hashing has attracted many research efforts. Nevertheless, backdoor attack, another famous threat to DNNs, has not been studied for deep hashing yet. Although various backdoor attacks have been proposed in the field of image classification, existing approaches failed to realize a truly imperceptive backdoor attack that enjoys invisible triggers and clean label setting simultaneously, and they also cannot meet the intrinsic demand of image retrieval backdoor. In this paper, we propose BadHash, the first generative-based imperceptible backdoor attack against deep hashing, which can effectively generate invisible and input-specific poisoned images with clean label. Specifically, we first propose a new conditional generative adversarial network (cGAN) pipeline to effectively generate poisoned samples. For any given benign image, it seeks to generate a natural-looking poisoned counterpart with a unique invisible trigger. In order to improve the attack effectiveness, we introduce a label-based contrastive learning network LabCLN to exploit the semantic characteristics of different labels, which are subsequently used for confusing and misleading the target model to learn the embedded trigger. We finally explore the mechanism of backdoor attacks on image retrieval in the hash space. Extensive experiments on multiple benchmark datasets verify that BadHash can generate imperceptible poisoned samples with strong attack ability and transferability over state-of-the-art deep hashing schemes.

Simple tabulation hashing dates back to Zobrist in 1970 and is defined as follows: Each key is viewed as \(c\) characters from some alphabet \(\Sigma\), we have \(c\) fully random hash functions \(h_0, \ldots, h_{c - 1} \colon \Sigma \to \{0, \ldots, 2^l - 1\}\), and a key \(x = (x_0, \ldots, x_{c - 1})\) is hashed to \(h(x) = h_0(x_0) \oplus \ldots \oplus h_{c - 1}(x_{c - 1})\) where \(\oplus\) is the bitwise XOR operation. The previous results on tabulation hashing by P{\v a}tra{\c s}cu and Thorup~[J.ACM’11] and by Aamand et al.~[STOC’20] focused on proving Chernoff-style tail bounds on hash-based sums, e.g., the number keys hashing to a given value, for simple tabulation hashing, but their bounds do not cover the entire tail. Chaoses are random variables of the form \(\sum a_{i_0, \ldots, i_{c - 1}} X_{i_0} \cdot \ldots \cdot X_{i_{c - 1}}\) where \(X_i\) are independent random variables. Chaoses are a well-studied concept from probability theory, and tight analysis has been proven in several instances, e.g., when the independent random variables are standard Gaussian variables and when the independent random variables have logarithmically convex tails. We notice that hash-based sums of simple tabulation hashing can be seen as a sum of chaoses that are not independent. This motivates us to use techniques from the theory of chaoses to analyze hash-based sums of simple tabulation hashing. In this paper, we obtain bounds for all the moments of hash-based sums for simple tabulation hashing which are tight up to constants depending only on \(c\). In contrast with the previous attempts, our approach will mostly be analytical and does not employ intricate combinatorial arguments. The improved analysis of simple tabulation hashing allows us to obtain bounds for the moments of hash-based sums for the mixed tabulation hashing introduced by Dahlgaard et al.~[FOCS’15].

Computational notebook software such as Jupyter Notebook is popular for data science tasks. Numerous computational notebooks are available on the Web and reusable; however, searching for computational notebooks manually is a tedious task, and so far, there are no tools to search for computational notebooks effectively and efficiently. In this paper, we propose a similarity search on computational notebooks and develop a new framework for the similarity search. Given contents (i.e., source codes, tabular data, libraries, and outputs formats) in computational notebooks as a query, the similarity search problem aims to find top-k computational notebooks with the most similar contents. We define two similarity measures; set-based and graph-based similarities. Set-based similarity handles each content independently, while graph-based similarity captures the relationships between contents. Our framework can effectively prune the candidates of computational notebooks that should not be in the top-k results. Furthermore, we develop optimization techniques such as caching and indexing to accelerate the search. Experiments using Kaggle notebooks show that our method, in particular graph-based similarity, can achieve high accuracy and high efficiency.

Deep hashing has been widely applied to large-scale image retrieval tasks owing to efficient computation and low storage cost by encoding high-dimensional image data into binary codes. Since binary codes do not contain as much information as float features, the essence of binary encoding is preserving the main context to guarantee retrieval quality. However, the existing hashing methods have great limitations on suppressing redundant background information and accurately encoding from Euclidean space to Hamming space by a simple sign function. In order to solve these problems, a Cross-Scale Context Extracted Hashing Network (CSCE-Net) is proposed in this paper. Firstly, we design a two-branch framework to capture fine-grained local information while maintaining high-level global semantic information. Besides, Attention guided Information Extraction module (AIE) is introduced between two branches, which suppresses areas of low context information cooperated with global sliding windows. Unlike previous methods, our CSCE-Net learns a content-related Dynamic Sign Function (DSF) to replace the original simple sign function. Therefore, the proposed CSCE-Net is context-sensitive and able to perform well on accurate image binary encoding. We further demonstrate that our CSCE-Net is superior to the existing hashing methods, which improves retrieval performance on standard benchmarks.

Ad-hoc search calls for the selection of appropriate answers from a massive-scale corpus. Nowadays, the embedding-based retrieval (EBR) becomes a promising solution, where deep learning based document representation and ANN search techniques are allied to handle this task. However, a major challenge is that the ANN index can be too large to fit into memory, given the considerable size of answer corpus. In this work, we tackle this problem with Bi-Granular Document Representation, where the lightweight sparse embeddings are indexed and standby in memory for coarse-grained candidate search, and the heavyweight dense embeddings are hosted in disk for fine-grained post verification. For the best of retrieval accuracy, a Progressive Optimization framework is designed. The sparse embeddings are learned ahead for high-quality search of candidates. Conditioned on the candidate distribution induced by the sparse embeddings, the dense embeddings are continuously learned to optimize the discrimination of ground-truth from the shortlisted candidates. Besides, two techniques: the contrastive quantization and the locality-centric sampling are introduced for the learning of sparse and dense embeddings, which substantially contribute to their performances. Thanks to the above features, our method effectively handles massive-scale EBR with strong advantages in accuracy: with up to +4.3% recall gain on million-scale corpus, and up to +17.5% recall gain on billion-scale corpus. Besides, Our method is applied to a major sponsored search platform with substantial gains on revenue (+1.95%), Recall (+1.01%) and CTR (+0.49%). Our code is available at https://github.com/microsoft/BiDR.

Most cloud services and distributed applications rely on hashing algorithms that allow dynamic scaling of a robust and efficient hash table. Examples include AWS, Google Cloud and BitTorrent. Consistent and rendezvous hashing are algorithms that minimize key remapping as the hash table resizes. While memory errors in large-scale cloud deployments are common, neither algorithm offers both efficiency and robustness. Hyperdimensional Computing is an emerging computational model that has inherent efficiency, robustness and is well suited for vector or hardware acceleration. We propose Hyperdimensional (HD) hashing and show that it has the efficiency to be deployed in large systems. Moreover, a realistic level of memory errors causes more than 20% mismatches for consistent hashing while HD hashing remains unaffected.

Transformer-based pre-trained language models are vocabulary-dependent, mapping by default each token to its corresponding embedding. This one-to-one mapping results into embedding matrices that occupy a lot of memory (i.e. millions of parameters) and grow linearly with the size of the vocabulary. Previous work on on-device transformers dynamically generate token embeddings on-the-fly without embedding matrices using locality-sensitive hashing over morphological information. These embeddings are subsequently fed into transformer layers for text classification. However, these methods are not pre-trained. Inspired by this line of work, we propose HashFormers, a new family of vocabulary-independent pre-trained transformers that support an unlimited vocabulary (i.e. all possible tokens in a corpus) given a substantially smaller fixed-sized embedding matrix. We achieve this by first introducing computationally cheap hashing functions that bucket together individual tokens to embeddings. We also propose three variants that do not require an embedding matrix at all, further reducing the memory requirements. We empirically demonstrate that HashFormers are more memory efficient compared to standard pre-trained transformers while achieving comparable predictive performance when fine-tuned on multiple text classification tasks. For example, our most efficient HashFormer variant has a negligible performance degradation (0.4\% on GLUE) using only 99.1K parameters for representing the embeddings compared to 12.3-38M parameters of state-of-the-art models.

Code search is to search reusable code snippets from source code corpus based on natural languages queries. Deep learning-based methods of code search have shown promising results. However, previous methods focus on retrieval accuracy but lacked attention to the efficiency of the retrieval process. We propose a novel method CoSHC to accelerate code search with deep hashing and code classification, aiming to perform an efficient code search without sacrificing too much accuracy. To evaluate the effectiveness of CoSHC, we apply our method to five code search models. Extensive experimental results indicate that compared with previous code search baselines, CoSHC can save more than 90% of retrieval time meanwhile preserving at least 99% of retrieval accuracy.

Unsupervised image hashing, which maps images into binary codes without supervision, is a compressor with a high compression rate. Hence, how to preserving meaningful information of the original data is a critical problem. Inspired by the large-scale vision pre-training model, known as ViT, which has shown significant progress for learning visual representations, in this paper, we propose a simple information-preserving compressor to finetune the ViT model for the target unsupervised hashing task. Specifically, from pixels to continuous features, we first propose a feature-preserving module, using the corrupted image as input to reconstruct the original feature from the pre-trained ViT model and the complete image, so that the feature extractor can focus on preserving the meaningful information of original data. Secondly, from continuous features to hash codes, we propose a hashing-preserving module, which aims to keep the semantic information from the pre-trained ViT model by using the proposed Kullback-Leibler divergence loss. Besides, the quantization loss and the similarity loss are added to minimize the quantization error. Our method is very simple and achieves a significantly higher degree of MAP on three benchmark image datasets.

Case-based Reasoning (CBR) on high-dimensional and heterogeneous data is a trending yet challenging and computationally expensive task in the real world. A promising approach is to obtain low-dimensional hash codes representing cases and perform a similarity retrieval of cases in Hamming space. However, previous methods based on data-independent hashing rely on random projections or manual construction, inapplicable to address specific data issues (e.g., high-dimensionality and heterogeneity) due to their insensitivity to data characteristics. To address these issues, this work introduces a novel deep hashing network to learn similarity-preserving compact hash codes for efficient case retrieval and proposes a deep-hashing-enabled CBR model HeCBR. Specifically, we introduce position embedding to represent heterogeneous features and utilize a multilinear interaction layer to obtain case embeddings, which effectively filtrates zero-valued features to tackle high-dimensionality and sparsity and captures inter-feature couplings. Then, we feed the case embeddings into fully-connected layers, and subsequently a hash layer generates hash codes with a quantization regularizer to control the quantization loss during relaxation. To cater to incremental learning of CBR, we further propose an adaptive learning strategy to update the hash function. Extensive experiments on public datasets show that HeCBR greatly reduces storage and significantly accelerates case retrieval. HeCBR achieves desirable performance compared with the state-of-the-art CBR methods and performs significantly better than hashing-based CBR methods in classification.

A key characteristic of deep recommendation models is the immense memory requirements of their embedding tables. These embedding tables can often reach hundreds of gigabytes which increases hardware requirements and training cost. A common technique to reduce model size is to hash all of the categorical variable identifiers (ids) into a smaller space. This hashing reduces the number of unique representations that must be stored in the embedding table; thus decreasing its size. However, this approach introduces collisions between semantically dissimilar ids that degrade model quality. We introduce an alternative approach, Learned Hash Functions, which instead learns a new mapping function that encourages collisions between semantically similar ids. We derive this learned mapping from historical data and embedding access patterns. We experiment with this technique on a production model and find that a mapping informed by the combination of access frequency and a learned low dimension embedding is the most effective. We demonstrate a small improvement relative to the hashing trick and other collision related compression techniques. This is ongoing work that explores the impact of categorical id collisions on recommendation model quality and how those collisions may be controlled to improve model performance.

In this work, we explore the relationship between free resolution of some monomial ideals and Generalized Hamming Weights (GHWs) of binary codes. More precisely, we look for a structure smaller than the set of codewords of minimal support that provides us some information about the GHWs. We prove that the first and second generalized Hamming weight of a binary linear code can be computed (by means of a graded free resolution) from a set of monomials associated to a binomial ideal related with the code. Moreover, the remaining weights are bounded by the Betti numbers for that set.

Existing Cross Modal Hashing (CMH) methods are mainly designed for balanced data, while imbalanced data with long-tail distribution is more general in real-world. Several long-tail hashing methods have been proposed but they can not adapt for multi-modal data, due to the complex interplay between labels and individuality and commonality information of multi-modal data. Furthermore, CMH methods mostly mine the commonality of multi-modal data to learn hash codes, which may override tail labels encoded by the individuality of respective modalities. In this paper, we propose LtCMH (Long-tail CMH) to handle imbalanced multi-modal data. LtCMH firstly adopts auto-encoders to mine the individuality and commonality of different modalities by minimizing the dependency between the individuality of respective modalities and by enhancing the commonality of these modalities. Then it dynamically combines the individuality and commonality with direct features extracted from respective modalities to create meta features that enrich the representation of tail labels, and binaries meta features to generate hash codes. LtCMH significantly outperforms state-of-the-art baselines on long-tail datasets and holds a better (or comparable) performance on datasets with balanced labels.

Hyperdimensional computing (HDC) is a paradigm for data representation and learning originating in computational neuroscience. HDC represents data as high-dimensional, low-precision vectors which can be used for a variety of information processing tasks like learning or recall. The mapping to high-dimensional space is a fundamental problem in HDC, and existing methods encounter scalability issues when the input data itself is high-dimensional. In this work, we explore a family of streaming encoding techniques based on hashing. We show formally that these methods enjoy comparable guarantees on performance for learning applications while being substantially more efficient than existing alternatives. We validate these results experimentally on a popular high-dimensional classification problem and show that our approach easily scales to very large data sets.

Image copy detection and retrieval from large databases leverage two components. First, a neural network maps an image to a vector representation, that is relatively robust to various transformations of the image. Second, an efficient but approximate similarity search algorithm trades scalability (size and speed) against quality of the search, thereby introducing a source of error. This paper improves the robustness of image copy detection with active indexing, that optimizes the interplay of these two components. We reduce the quantization loss of a given image representation by making imperceptible changes to the image before its release. The loss is back-propagated through the deep neural network back to the image, under perceptual constraints. These modifications make the image more retrievable. Our experiments show that the retrieval and copy detection of activated images is significantly improved. For instance, activation improves by \(+40\%\) the Recall1@1 on various image transformations, and for several popular indexing structures based on product quantization and locality sensitivity hashing.

Graph neural networks (GNNs) are deep learning models designed specifically for graph data, and they typically rely on node features as the input to the first layer. When applying such a type of network on the graph without node features, one can extract simple graph-based node features (e.g., number of degrees) or learn the input node representations (i.e., embeddings) when training the network. While the latter approach, which trains node embeddings, more likely leads to better performance, the number of parameters associated with the embeddings grows linearly with the number of nodes. It is therefore impractical to train the input node embeddings together with GNNs within graphics processing unit (GPU) memory in an end-to-end fashion when dealing with industrial-scale graph data. Inspired by the embedding compression methods developed for natural language processing (NLP) tasks, we develop a node embedding compression method where each node is compactly represented with a bit vector instead of a floating-point vector. The parameters utilized in the compression method can be trained together with GNNs. We show that the proposed node embedding compression method achieves superior performance compared to the alternatives.

Person Re-Identification (ReID) matches pedestrians across disjoint cameras. Existing ReID methods adopting real-value feature descriptors have achieved high accuracy, but they are low in efficiency due to the slow Euclidean distance computation as well as complex quick-sort algorithms. Recently, some works propose to yield binary encoded person descriptors which instead only require fast Hamming distance computation and simple counting-sort algorithms. However, the performances of such binary encoded descriptors, especially with short code (e.g., 32 and 64 bits), are hardly satisfactory given the sparse binary space. To strike a balance between the model accuracy and efficiency, we propose a novel Sub-space Consistency Regularization (SCR) algorithm that can speed up the ReID procedure by \(0.25\) times than real-value features under the same dimensions whilst maintaining a competitive accuracy, especially under short codes. SCR transforms real-value features vector (e.g., 2048 float32) with short binary codes (e.g., 64 bits) by first dividing real-value features vector into \(M\) sub-spaces, each with \(C\) clustered centroids. Thus the distance between two samples can be expressed as the summation of the respective distance to the centroids, which can be sped up by offline calculation and maintained via a look-up table. On the other side, these real-value centroids help to achieve significantly higher accuracy than using binary code. Lastly, we convert the distance look-up table to be integer and apply the counting-sort algorithm to speed up the ranking stage. We also propose a novel consistency regularization with an iterative framework. Experimental results on Market-1501 and DukeMTMC-reID show promising and exciting results. Under short code, our proposed SCR enjoys Real-value-level accuracy and Hashing-level speed.

Candidate retrieval is the first stage in recommendation systems, where a light-weight system is used to retrieve potentially relevant items for an input user. These candidate items are then ranked and pruned in later stages of recommender systems using a more complex ranking model. As the top of the recommendation funnel, it is important to retrieve a high-recall candidate set to feed into downstream ranking models. A common approach is to leverage approximate nearest neighbor (ANN) search from a single dense query embedding; however, this approach this can yield a low-diversity result set with many near duplicates. As users often have multiple interests, candidate retrieval should ideally return a diverse set of candidates reflective of the user’s multiple interests. To this end, we introduce kNN-Embed, a general approach to improving diversity in dense ANN-based retrieval. kNN-Embed represents each user as a smoothed mixture over learned item clusters that represent distinct “interests” of the user. By querying each of a user’s mixture component in proportion to their mixture weights, we retrieve a high-diversity set of candidates reflecting elements from each of a user’s interests. We experimentally compare kNN-Embed to standard ANN candidate retrieval, and show significant improvements in overall recall and improved diversity across three datasets. Accompanying this work, we open source a large Twitter follow-graph dataset (https://huggingface.co/datasets/Twitter/TwitterFollowGraph), to spur further research in graph-mining and representation learning for recommender systems.

Convolutional Neural Networks (CNNs) are known for requiring extensive computational resources, and quantization is among the best and most common methods for compressing them. While aggressive quantization (i.e., less than 4-bits) performs well for classification, it may cause severe performance degradation in image-to-image tasks such as semantic segmentation and depth estimation. In this paper, we propose Wavelet Compressed Convolution (WCC)—a novel approach for high-resolution activation maps compression integrated with point-wise convolutions, which are the main computational cost of modern architectures. To this end, we use an efficient and hardware-friendly Haar-wavelet transform, known for its effectiveness in image compression, and define the convolution on the compressed activation map. We experiment with various tasks that benefit from high-resolution input. By combining WCC with light quantization, we achieve compression rates equivalent to 1-4bit activation quantization with relatively small and much more graceful degradation in performance. Our code is available at https://github.com/BGUCompSci/WaveletCompressedConvolution.

Image hashing is a principled approximate nearest neighbor approach to find similar items to a query in a large collection of images. Hashing aims to learn a binary-output function that maps an image to a binary vector. For optimal retrieval performance, producing balanced hash codes with low-quantization error to bridge the gap between the learning stage’s continuous relaxation and the inference stage’s discrete quantization is important. However, in the existing deep supervised hashing methods, coding balance and low-quantization error are difficult to achieve and involve several losses. We argue that this is because the existing quantization approaches in these methods are heuristically constructed and not effective to achieve these objectives. This paper considers an alternative approach to learning the quantization constraints. The task of learning balanced codes with low quantization error is re-formulated as matching the learned distribution of the continuous codes to a pre-defined discrete, uniform distribution. This is equivalent to minimizing the distance between two distributions. We then propose a computationally efficient distributional distance by leveraging the discrete property of the hash functions. This distributional distance is a valid distance and enjoys lower time and sample complexities. The proposed single-loss quantization objective can be integrated into any existing supervised hashing method to improve code balance and quantization error. Experiments confirm that the proposed approach substantially improves the performance of several representative hashing~methods.

Fast item ranking is an important task in recommender systems. In previous works, graph-based Approximate Nearest Neighbor (ANN) approaches have demonstrated good performance on item ranking tasks with generic searching/matching measures (including complex measures such as neural network measures). However, since these ANN approaches must go through the neural measures several times during ranking, the computation is not practical if the neural measure is a large network. On the other hand, fast item ranking using existing hashing-based approaches, such as Locality Sensitive Hashing (LSH), only works with a limited set of measures. Previous learning-to-hash approaches are also not suitable to solve the fast item ranking problem since they can take a significant amount of time and computation to train the hash functions. Hashing approaches, however, are attractive because they provide a principle and efficient way to retrieve candidate items. In this paper, we propose a simple and effective learning-to-hash approach for the fast item ranking problem that can be used for any type of measure, including neural network measures. Specifically, we solve this problem with an asymmetric hashing framework based on discrete inner product fitting. We learn a pair of related hash functions that map heterogeneous objects (e.g., users and items) into a common discrete space where the inner product of their binary codes reveals their true similarity defined via the original searching measure. The fast ranking problem is reduced to an ANN search via this asymmetric hashing scheme. Then, we propose a sampling strategy to efficiently select relevant and contrastive samples to train the hashing model. We empirically validate the proposed method against the existing state-of-the-art fast item ranking methods in several combinations of non-linear searching functions and prominent datasets.

Leveraging supervised information can lead to superior retrieval performance in the image hashing domain but the performance degrades significantly without enough labeled data. One effective solution to boost performance is to employ generative models, such as Generative Adversarial Networks (GANs), to generate synthetic data in an image hashing model. However, GAN-based methods are difficult to train, which prevents the hashing approaches from jointly training the generative models and the hash functions. This limitation results in sub-optimal retrieval performance. To overcome this limitation, we propose a novel framework, the generative cooperative hashing network, which is based on energy-based cooperative learning. This framework jointly learns a powerful generative representation of the data and a robust hash function via two components: a top-down contrastive pair generator that synthesizes contrastive images and a bottom-up multipurpose descriptor that simultaneously represents the images from multiple perspectives, including probability density, hash code, latent code, and category. The two components are jointly learned via a novel likelihood-based cooperative learning scheme. We conduct experiments on several real-world datasets and show that the proposed method outperforms the competing hashing supervised methods, achieving up to 10\% relative improvement over the current state-of-the-art supervised hashing methods, and exhibits a significantly better performance in out-of-distribution retrieval.

Hashing technology gains much attention in protecting the biometric template lately. For instance, Index-of-Max (IoM), a recent reported hashing technique, is a ranking-based locality sensitive hashing technique, which illustrates the feasibility to protect the ordered and fixed-length biometric template. However, biometric templates are not always in the form of ordered and fixed-length, rather it may be an unordered and variable size point set e.g. fingerprint minutiae, which restricts the usage of the traditional hashing technology. In this paper, we proposed a generalized version of IoM hashing namely gIoM, and therefore the unordered and variable size biometric template can be used. We demonstrate a realization using a well-known variable size feature vector, fingerprint Minutia Cylinder-Code (MCC). The gIoM transforms MCC into index domain to form indexing-based feature representation. Consequently, the inversion of MCC from the transformed representation is computational infeasible, thus to achieve non-invertibility while the performance is preserved. Public fingerprint databases FVC2002 and FVC2004 are employed for experiment as benchmark to demonstrate a fair comparison with other methods. Moreover, the security and privacy analysis suggest that gIoM meets the criteria of template protection: non-invertibility, revocability, and non-linkability.

This paper presents a deep learning approach for image retrieval and pattern spotting in digital collections of historical documents. First, a region proposal algorithm detects object candidates in the document page images. Next, deep learning models are used for feature extraction, considering two distinct variants, which provide either real-valued or binary code representations. Finally, candidate images are ranked by computing the feature similarity with a given input query. A robust experimental protocol evaluates the proposed approach considering each representation scheme (real-valued and binary code) on the DocExplore image database. The experimental results show that the proposed deep models compare favorably to the state-of-the-art image retrieval approaches for images of historical documents, outperforming other deep models by 2.56 percentage points using the same techniques for pattern spotting. Besides, the proposed approach also reduces the search time by up to 200x and the storage cost up to 6,000x when compared to related works based on real-valued representations.

We show that a randomly chosen linear map over a finite field gives a good hash function in the \(\ell_\infty\) sense. More concretely, consider a set \(S \subset \mathbb{F}q^n\) and a randomly chosen linear map \(L : \mathbb{F}_q^n \to \mathbb{F}_q^t\) with \(q^t\) taken to be sufficiently smaller than \( |S|\). Let \(U_S\) denote a random variable distributed uniformly on \(S\). Our main theorem shows that, with high probability over the choice of \(L\), the random variable \(L(U_S)\) is close to uniform in the \(\ell\infty\) norm. In other words, {\em every} element in the range \(\mathbb{F}_q^t\) has about the same number of elements in \(S\) mapped to it. This complements the widely-used Leftover Hash Lemma (LHL) which proves the analog statement under the statistical, or \(\ell_1\), distance (for a richer class of functions) as well as prior work on the expected largest ‘bucket size’ in linear hash functions [ADMPT99]. By known bounds from the load balancing literature [RS98], our results are tight and show that linear functions hash as well as trully random function up to a constant factor in the entropy loss. Our proof leverages a connection between linear hashing and the finite field Kakeya problem and extends some of the tools developed in this area, in particular the polynomial method.

The development of unsupervised hashing is advanced by the recent popular contrastive learning paradigm. However, previous contrastive learning-based works have been hampered by (1) insufficient data similarity mining based on global-only image representations, and (2) the hash code semantic loss caused by the data augmentation. In this paper, we propose a novel method, namely Weighted Contrative Hashing (WCH), to take a step towards solving these two problems. We introduce a novel mutual attention module to alleviate the problem of information asymmetry in network features caused by the missing image structure during contrative augmentation. Furthermore, we explore the fine-grained semantic relations between images, i.e., we divide the images into multiple patches and calculate similarities between patches. The aggregated weighted similarities, which reflect the deep image relations, are distilled to facilitate the hash codes learning with a distillation loss, so as to obtain better retrieval performance. Extensive experiments show that the proposed WCH significantly outperforms existing unsupervised hashing methods on three benchmark datasets.

Learning to hash pictures a list-wise sorting problem. Its testing metrics, e.g., mean-average precision, count on a sorted candidate list ordered by pair-wise code similarity. However, scarcely does one train a deep hashing model with the sorted results end-to-end because of the non-differentiable nature of the sorting operation. This inconsistency in the objectives of training and test may lead to sub-optimal performance since the training loss often fails to reflect the actual retrieval metric. In this paper, we tackle this problem by introducing Naturally-Sorted Hashing (NSH). We sort the Hamming distances of samples’ hash codes and accordingly gather their latent representations for self-supervised training. Thanks to the recent advances in differentiable sorting approximations, the hash head receives gradients from the sorter so that the hash encoder can be optimized along with the training procedure. Additionally, we describe a novel Sorted Noise-Contrastive Estimation (SortedNCE) loss that selectively picks positive and negative samples for contrastive learning, which allows NSH to mine data semantic relations during training in an unsupervised manner. Our extensive experiments show the proposed NSH model significantly outperforms the existing unsupervised hashing methods on three benchmarked datasets.

We define new families of Tillich-Z'emor hash functions, using higher dimensional special linear groups over finite fields as platforms. The Cayley graphs of these groups combine fast mixing properties and high girth, which together give rise to good preimage and collision resistance of the corresponding hash functions. We justify the claim that the resulting hash functions are post-quantum secure.

Deep hashing has been widely applied in large-scale data retrieval due to its superior retrieval efficiency and low storage cost. However, data are often scattered in data silos with privacy concerns, so performing centralized data storage and retrieval is not always possible. Leveraging the concept of federated learning (FL) to perform deep hashing is a recent research trend. However, existing frameworks mostly rely on the aggregation of the local deep hashing models, which are trained by performing similarity learning with local skewed data only. Therefore, they cannot work well for non-IID clients in a real federated environment. To overcome these challenges, we propose a novel federated hashing framework that enables participating clients to jointly train the shared deep hashing model by leveraging the prototypical hash codes for each class. Globally, the transmission of global prototypes with only one prototypical hash code per class will minimize the impact of communication cost and privacy risk. Locally, the use of global prototypes are maximized by jointly training a discriminator network and the local hashing network. Extensive experiments on benchmark datasets are conducted to demonstrate that our method can significantly improve the performance of the deep hashing model in the federated environments with non-IID data distributions.

Embedding tables are used by machine learning systems to work with categorical features. In modern Recommendation Systems, these tables can be very large, necessitating the development of new methods for fitting them in memory, even during training. We suggest Clustered Compositional Embeddings (CCE) which combines clustering-based compression like quantization to codebooks with dynamic methods like The Hashing Trick and Compositional Embeddings (Shi et al., 2020). Experimentally CCE achieves the best of both worlds: The high compression rate of codebook-based quantization, but dynamically like hashing-based methods, so it can be used during training. Theoretically, we prove that CCE is guaranteed to converge to the optimal codebook and give a tight bound for the number of iterations required.

To make full use of computer vision technology in stores, it is required to consider the actual needs that fit the characteristics of the retail scene. Pursuing this goal, we introduce the United Retail Datasets (Unitail), a large-scale benchmark of basic visual tasks on products that challenges algorithms for detecting, reading, and matching. With 1.8M quadrilateral-shaped instances annotated, the Unitail offers a detection dataset to align product appearance better. Furthermore, it provides a gallery-style OCR dataset containing 1454 product categories, 30k text regions, and 21k transcriptions to enable robust reading on products and motivate enhanced product matching. Besides benchmarking the datasets using various state-of-the-arts, we customize a new detector for product detection and provide a simple OCR-based matching solution that verifies its effectiveness.

Approximate K-Nearest Neighbor Search (AKNNS) has now become ubiquitous in modern applications, for example, as a fast search procedure with two tower deep learning models. Graph-based methods for AKNNS in particular have received great attention due to their superior performance. These methods rely on greedy graph search to traverse the data points as embedding vectors in a database. Under this greedy search scheme, we make a key observation: many distance computations do not influence search updates so these computations can be approximated without hurting performance. As a result, we propose FINGER, a fast inference method to achieve efficient graph search. FINGER approximates the distance function by estimating angles between neighboring residual vectors with low-rank bases and distribution matching. The approximated distance can be used to bypass unnecessary computations, which leads to faster searches. Empirically, accelerating a popular graph-based method named HNSW by FINGER is shown to outperform existing graph-based methods by 20%-60% across different benchmark datasets.

Model-based methods for recommender systems have been studied extensively for years. Modern recommender systems usually resort to 1) representation learning models which define user-item preference as the distance between their embedding representations, and 2) embedding-based Approximate Nearest Neighbor (ANN) search to tackle the efficiency problem introduced by large-scale corpus. While providing efficient retrieval, the embedding-based retrieval pattern also limits the model capacity since the form of user-item preference measure is restricted to the distance between their embedding representations. However, for other more precise user-item preference measures, e.g., preference scores directly derived from a deep neural network, they are computationally intractable because of the lack of an efficient retrieval method, and an exhaustive search for all user-item pairs is impractical. In this paper, we propose a novel method to extend ANN search to arbitrary matching functions, e.g., a deep neural network. Our main idea is to perform a greedy walk with a matching function in a similarity graph constructed from all items. To solve the problem that the similarity measures of graph construction and user-item matching function are heterogeneous, we propose a pluggable adversarial training task to ensure the graph search with arbitrary matching function can achieve fairly high precision. Experimental results in both open source and industry datasets demonstrate the effectiveness of our method. The proposed method has been fully deployed in the Taobao display advertising platform and brings a considerable advertising revenue increase. We also summarize our detailed experiences in deployment in this paper.

Learning vectorized embeddings is at the core of various recommender systems for user-item matching. To perform efficient online inference, representation quantization, aiming to embed the latent features by a compact sequence of discrete numbers, recently shows the promising potentiality in optimizing both memory and computation overheads. However, existing work merely focuses on numerical quantization whilst ignoring the concomitant information loss issue, which, consequently, leads to conspicuous performance degradation. In this paper, we propose a novel quantization framework to learn Binarized Graph Representations for Top-K Recommendation (BiGeaR). BiGeaR introduces multi-faceted quantization reinforcement at the pre-, mid-, and post-stage of binarized representation learning, which substantially retains the representation informativeness against embedding binarization. In addition to saving the memory footprint, BiGeaR further develops solid online inference acceleration with bitwise operations, providing alternative flexibility for the realistic deployment. The empirical results over five large real-world benchmarks show that BiGeaR achieves about 22%~40% performance improvement over the state-of-the-art quantization-based recommender system, and recovers about 95%~102% of the performance capability of the best full-precision counterpart with over 8x time and space reduction.

Many bioinformatics applications involve bucketing a set of sequences where each sequence is allowed to be assigned into multiple buckets. To achieve both high sensitivity and precision, bucketing methods are desired to assign similar sequences into the same bucket while assigning dissimilar sequences into distinct buckets. Existing \(k\)-mer-based bucketing methods have been efficient in processing sequencing data with low error rate, but encounter much reduced sensitivity on data with high error rate. Locality-sensitive hashing (LSH) schemes are able to mitigate this issue through tolerating the edits in similar sequences, but state-of-the-art methods still have large gaps. Here we generalize the LSH function by allowing it to hash one sequence into multiple buckets. Formally, a bucketing function, which maps a sequence (of fixed length) into a subset of buckets, is defined to be \((d_1, d_2)\)-sensitive if any two sequences within an edit distance of \(d_1\) are mapped into at least one shared bucket, and any two sequences with distance at least \(d_2\) are mapped into disjoint subsets of buckets. We construct locality-sensitive bucketing (LSB) functions with a variety of values of \((d_1,d_2)\) and analyze their efficiency with respect to the total number of buckets needed as well as the number of buckets that a specific sequence is mapped to. We also prove lower bounds of these two parameters in different settings and show that some of our constructed LSB functions are optimal. These results provide theoretical foundations for their practical use in analyzing sequences with high error rate while also providing insights for the hardness of designing ungapped LSH functions.

A minimal perfect hash function (MPHF) bijectively maps a set S of objects to the first |S| integers. It can be used as a building block in databases and data compression. RecSplit [Esposito et al., ALENEX’20] is currently the most space efficient practical minimal perfect hash function. It heavily relies on trying out hash functions in a brute force way. We introduce rotation fitting, a new technique that makes the search more efficient by drastically reducing the number of tried hash functions. Additionally, we greatly improve the construction time of RecSplit by harnessing parallelism on the level of bits, vectors, cores, and GPUs. In combination, the resulting improvements yield speedups up to 239 on an 8-core CPU and up to 5438 using a GPU. The original single-threaded RecSplit implementation needs 1.5 hours to construct an MPHF for 5 Million objects with 1.56 bits per object. On the GPU, we achieve the same space usage in just 5 seconds. Given that the speedups are larger than the increase in energy consumption, our implementation is more energy efficient than the original implementation.

Indexing is an effective way to support efficient query processing in large databases. Recently the concept of learned index, which replaces or complements traditional index structures with machine learning models, has been actively explored to reduce storage and search costs. However, accurate and efficient similarity query processing in high-dimensional metric spaces remains to be an open challenge. In this paper, we propose a novel indexing approach called LIMS that uses data clustering, pivot-based data transformation techniques and learned indexes to support efficient similarity query processing in metric spaces. In LIMS, the underlying data is partitioned into clusters such that each cluster follows a relatively uniform data distribution. Data redistribution is achieved by utilizing a small number of pivots for each cluster. Similar data are mapped into compact regions and the mapped values are totally ordinal. Machine learning models are developed to approximate the position of each data record on disk. Efficient algorithms are designed for processing range queries and nearest neighbor queries based on LIMS, and for index maintenance with dynamic updates. Extensive experiments on real-world and synthetic datasets demonstrate the superiority of LIMS compared with traditional indexes and state-of-the-art learned indexes.

We address the maximum size of binary codes and binary constant weight codes with few distances. Previous works established a number of bounds for these quantities as well as the exact values for a range of small code lengths. As our main results, we determine the exact size of maximal binary codes with two distances for all lengths \(n\ge 6\) as well as the exact size of maximal binary constant weight codes with 2,3, and 4 distances for several values of the weight and for all but small lengths.

The monotone minimal perfect hash function (MMPHF) problem is the following indexing problem. Given a set \(S= \{s_1,\ldots,s_n\}\) of \(n\) distinct keys from a universe \(U\) of size \(u\), create a data structure \(DS\) that answers the following query: [ RankOp(q) = \text{rank of } q \text{ in } S \text{ for all } q\in S ~\text{ and arbitrary answer otherwise.} ] Solutions to the MMPHF problem are in widespread use in both theory and practice. The best upper bound known for the problem encodes \(DS\) in \(O(nlogloglog u)\) bits and performs queries in \(O(log u)\) time. It has been an open problem to either improve the space upper bound or to show that this somewhat odd looking bound is tight. In this paper, we show the latter: specifically that any data structure (deterministic or randomized) for monotone minimal perfect hashing of any collection of \(n\) elements from a universe of size \(u\) requires \(Ω(n \cdot logloglog{u})\) expected bits to answer every query correctly. We achieve our lower bound by defining a graph \(\mathbf{G}\) where the nodes are the possible \({u \choose n}\) inputs and where two nodes are adjacent if they cannot share the same \(DS\). The size of \(DS\) is then lower bounded by the log of the chromatic number of \(\mathbf{G}\). Finally, we show that the fractional chromatic number (and hence the chromatic number) of \(\mathbf{G}\) is lower bounded by \(2^{Ω(n logloglog u)}\).

The growing operational capability of global Earth Observation (EO) creates new opportunities for data-driven approaches to understand and protect our planet. However, the current use of EO archives is very restricted due to the huge archive sizes and the limited exploration capabilities provided by EO platforms. To address this limitation, we have recently proposed MiLaN, a content-based image retrieval approach for fast similarity search in satellite image archives. MiLaN is a deep hashing network based on metric learning that encodes high-dimensional image features into compact binary hash codes. We use these codes as keys in a hash table to enable real-time nearest neighbor search and highly accurate retrieval. In this demonstration, we showcase the efficiency of MiLaN by integrating it with EarthQube, a browser and search engine within AgoraEO. EarthQube supports interactive visual exploration and Query-by-Example over satellite image repositories. Demo visitors will interact with EarthQube playing the role of different users that search images in a large-scale remote sensing archive by their semantic content and apply other filters.

The amount of data stored in data repositories increases every year. This makes it challenging to link records between different datasets across companies and even internally, while adhering to privacy regulations. Address or name changes, and even different spelling used for entity data, can prevent companies from using private deduplication or record-linking solutions such as private set intersection (PSI). To this end, we propose a new and efficient privacy-preserving record linkage (PPRL) protocol that combines PSI and local sensitive hash (LSH) functions, and runs in linear time. We explain the privacy guarantees that our protocol provides and demonstrate its practicality by executing the protocol over two datasets with \(2^{20}\) records each, in \(11-45\) minutes, depending on network settings.

Recently, to improve the unsupervised image retrieval performance, plenty of unsupervised hashing methods have been proposed by designing a semantic similarity matrix, which is based on the similarities between image features extracted by a pre-trained CNN model. However, most of these methods tend to ignore high-level abstract semantic concepts contained in images. Intuitively, concepts play an important role in calculating the similarity among images. In real-world scenarios, each image is associated with some concepts, and the similarity between two images will be larger if they share more identical concepts. Inspired by the above intuition, in this work, we propose a novel Unsupervised Hashing with Semantic Concept Mining, called UHSCM, which leverages a VLP model to construct a high-quality similarity matrix. Specifically, a set of randomly chosen concepts is first collected. Then, by employing a vision-language pretraining (VLP) model with the prompt engineering which has shown strong power in visual representation learning, the set of concepts is denoised according to the training images. Next, the proposed method UHSCM applies the VLP model with prompting again to mine the concept distribution of each image and construct a high-quality semantic similarity matrix based on the mined concept distributions. Finally, with the semantic similarity matrix as guiding information, a novel hashing loss with a modified contrastive loss based regularization item is proposed to optimize the hashing network. Extensive experiments on three benchmark datasets show that the proposed method outperforms the state-of-the-art baselines in the image retrieval task.

Recent research has sought to accelerate cryptographic hash functions as they are at the core of modern cryptography. Traditional designs, however, suffer from the von Neumann bottleneck that originates from the separation of processing and memory units. An emerging solution to overcome this bottleneck is processing-in-memory (PIM): performing logic within the same devices responsible for memory to eliminate data-transfer and simultaneously provide massive computational parallelism. In this paper, we seek to vastly accelerate the state-of-the-art SHA-3 cryptographic function using the memristive memory processing unit (mMPU), a general-purpose memristive PIM architecture. To that end, we propose a novel in-memory algorithm for variable rotation, and utilize an efficient mapping of the SHA-3 state vector for memristive crossbar arrays to efficiently exploit PIM parallelism. We demonstrate a massive energy efficiency of 1,422 Gbps/W, improving a state-of-the-art memristive SHA-3 accelerator (SHINE-2) by 4.6x.

Similarity search based on a distance function in metric spaces is a fundamental problem for many applications. Queries for similar objects lead to the well-known machine learning task of nearest-neighbours identification. Many data indexing strategies, collectively known as Metric Access Methods (MAM), have been proposed to speed up queries for similar elements in this context. Moreover, since exact approaches to solve similarity queries can be complex and time-consuming, alternative options have appeared to reduce query execution time, such as returning approximate results or resorting to distributed computing platforms. In this paper, we introduce MASK (Multilevel Approximate Similarity search with \(k\)-means), an unconventional application of the \(k\)-means algorithm as the foundation of a multilevel index structure for approximate similarity search, suitable for metric spaces. We show that inherent properties of \(k\)-means, like representing high-density data areas with fewer prototypes, can be leveraged for this purpose. An implementation of this new indexing method is evaluated, using a synthetic dataset and a real-world dataset in a high-dimensional and high-sparsity space. Results are promising and underpin the applicability of this novel indexing method in multiple domains.

With the recent boom of video-based social platforms (e.g., YouTube and TikTok), video retrieval using sentence queries has become an important demand and attracts increasing research attention. Despite the decent performance, existing text-video retrieval models in vision and language communities are impractical for large-scale Web search because they adopt brute-force search based on high-dimensional embeddings. To improve efficiency, Web search engines widely apply vector compression libraries (e.g., FAISS) to post-process the learned embeddings. Unfortunately, separate compression from feature encoding degrades the robustness of representations and incurs performance decay. To pursue a better balance between performance and efficiency, we propose the first quantized representation learning method for cross-view video retrieval, namely Hybrid Contrastive Quantization (HCQ). Specifically, HCQ learns both coarse-grained and fine-grained quantizations with transformers, which provide complementary understandings for texts and videos and preserve comprehensive semantic information. By performing Asymmetric-Quantized Contrastive Learning (AQ-CL) across views, HCQ aligns texts and videos at coarse-grained and multiple fine-grained levels. This hybrid-grained learning strategy serves as strong supervision on the cross-view video quantization model, where contrastive learning at different levels can be mutually promoted. Extensive experiments on three Web video benchmark datasets demonstrate that HCQ achieves competitive performance with state-of-the-art non-compressed retrieval methods while showing high efficiency in storage and computation. Code and configurations are available at https://github.com/gimpong/WWW22-HCQ.

Self-Supervised Video Hashing (SSVH) models learn to generate short binary representations for videos without ground-truth supervision, facilitating large-scale video retrieval efficiency and attracting increasing research attention. The success of SSVH lies in the understanding of video content and the ability to capture the semantic relation among unlabeled videos. Typically, state-of-the-art SSVH methods consider these two points in a two-stage training pipeline, where they firstly train an auxiliary network by instance-wise mask-and-predict tasks and secondly train a hashing model to preserve the pseudo-neighborhood structure transferred from the auxiliary network. This consecutive training strategy is inflexible and also unnecessary. In this paper, we propose a simple yet effective one-stage SSVH method called ConMH, which incorporates video semantic information and video similarity relationship understanding in a single stage. To capture video semantic information for better hashing learning, we adopt an encoder-decoder structure to reconstruct the video from its temporal-masked frames. Particularly, we find that a higher masking ratio helps video understanding. Besides, we fully exploit the similarity relationship between videos by maximizing agreement between two augmented views of a video, which contributes to more discriminative and robust hash codes. Extensive experiments on three large-scale video datasets (i.e., FCVID, ActivityNet and YFCC) indicate that ConMH achieves state-of-the-art results. Code is available at https://github.com/huangmozhi9527/ConMH.

Cross-modal hashing still has some challenges needed to address: (1) most existing CMH methods take graphs as input to model data distribution. These methods omit to consider the correlation of graph structure among multiple modalities; (2) most existing CMH methods ignores considering the fusion affinity among multi-modalities data; (3) most existing CMH methods relax the discrete constraints to solve the optimization objective, significantly degrading the retrieval performance. To solve the above limitations, we propose a novel Anchor Graph Structure Fusion Hashing (AGSFH). AGSFH constructs the anchor graph structure fusion matrix from different anchor graphs of multiple modalities with the Hadamard product, which can fully exploit the geometric property of underlying data structure. Based on the anchor graph structure fusion matrix, AGSFH attempts to directly learn an intrinsic anchor graph, where the structure of the intrinsic anchor graph is adaptively tuned so that the number of components of the intrinsic graph is exactly equal to the number of clusters. Besides, AGSFH preserves the anchor fusion affinity into the common binary Hamming space. Furthermore, a discrete optimization framework is designed to learn the unified binary codes. Extensive experimental results on three public social datasets demonstrate the superiority of AGSFH.

SNS providers are known to carry out the recompression and resizing of uploaded videos/images, but most conventional methods for detecting tampered videos/images are not robust enough against such operations. In addition, videos are temporally operated such as the insertion of new frames and the permutation of frames, of which operations are difficult to be detected by using conventional methods. Accordingly, in this paper, we propose a novel method with a robust hashing algorithm for detecting temporally operated videos even when applying resizing and compression to the videos.

Deep hashing has been extensively utilized in massive image retrieval because of its efficiency and effectiveness. However, deep hashing models are vulnerable to adversarial examples, making it essential to develop adversarial defense methods for image retrieval. Existing solutions achieved limited defense performance because of using weak adversarial samples for training and lacking discriminative optimization objectives to learn robust features. In this paper, we present a min-max based Center-guided Adversarial Training, namely CgAT, to improve the robustness of deep hashing networks through worst adversarial examples. Specifically, we first formulate the center code as a semantically-discriminative representative of the input image content, which preserves the semantic similarity with positive samples and dissimilarity with negative examples. We prove that a mathematical formula can calculate the center code immediately. After obtaining the center codes in each optimization iteration of the deep hashing network, they are adopted to guide the adversarial training process. On the one hand, CgAT generates the worst adversarial examples as augmented data by maximizing the Hamming distance between the hash codes of the adversarial examples and the center codes. On the other hand, CgAT learns to mitigate the effects of adversarial samples by minimizing the Hamming distance to the center codes. Extensive experiments on the benchmark datasets demonstrate the effectiveness of our adversarial training algorithm in defending against adversarial attacks for deep hashing-based retrieval. Compared with the current state-of-the-art defense method, we significantly improve the defense performance by an average of 18.61\%, 12.35\%, and 11.56\% on FLICKR-25K, NUS-WIDE, and MS-COCO, respectively. The code is available at https://github.com/xunguangwang/CgAT.

This paper addresses the construction of inverted index for large-scale image retrieval. The inverted index proposed by J. Sivic brings a significant acceleration by reducing distance computations with only a small fraction of the database. The state-of-the-art inverted indices aim to build finer partitions that produce a concise and accurate candidate list. However, partitioning in these frameworks is generally achieved by unsupervised clustering methods which ignore the semantic information of images. In this paper, we replace the clustering method with image classification, during the construction of codebook. We then propose a merging and splitting method to solve the problem that the number of partitions is unchangeable in the inverted semantic-index. Next, we combine our semantic-index with the product quantization (PQ) so as to alleviate the accuracy loss caused by PQ compression. Finally, we evaluate our model on large-scale image retrieval benchmarks. Experiment results demonstrate that our model can significantly improve the retrieval accuracy by generating high-quality candidate lists.

Unsupervised hashing has attracted much attention for binary representation learning due to the requirement of economical storage and efficiency of binary codes. It aims to encode high-dimensional features in the Hamming space with similarity preservation between instances. However, most existing methods learn hash functions in manifold-based approaches. Those methods capture the local geometric structures (i.e., pairwise relationships) of data, and lack satisfactory performance in dealing with real-world scenarios that produce similar features (e.g. color and shape) with different semantic information. To address this challenge, in this work, we propose an effective unsupervised method, namely Jointly Personalized Sparse Hashing (JPSH), for binary representation learning. To be specific, firstly, we propose a novel personalized hashing module, i.e., Personalized Sparse Hashing (PSH). Different personalized subspaces are constructed to reflect category-specific attributes for different clusters, adaptively mapping instances within the same cluster to the same Hamming space. In addition, we deploy sparse constraints for different personalized subspaces to select important features. We also collect the strengths of the other clusters to build the PSH module with avoiding over-fitting. Then, to simultaneously preserve semantic and pairwise similarities in our JPSH, we incorporate the PSH and manifold-based hash learning into the seamless formulation. As such, JPSH not only distinguishes the instances from different clusters, but also preserves local neighborhood structures within the cluster. Finally, an alternating optimization algorithm is adopted to iteratively capture analytical solutions of the JPSH model. Extensive experiments on four benchmark datasets verify that the JPSH outperforms several hashing algorithms on the similarity search task.

Neural graphics primitives, parameterized by fully connected neural networks, can be costly to train and evaluate. We reduce this cost with a versatile new input encoding that permits the use of a smaller network without sacrificing quality, thus significantly reducing the number of floating point and memory access operations: a small neural network is augmented by a multiresolution hash table of trainable feature vectors whose values are optimized through stochastic gradient descent. The multiresolution structure allows the network to disambiguate hash collisions, making for a simple architecture that is trivial to parallelize on modern GPUs. We leverage this parallelism by implementing the whole system using fully-fused CUDA kernels with a focus on minimizing wasted bandwidth and compute operations. We achieve a combined speedup of several orders of magnitude, enabling training of high-quality neural graphics primitives in a matter of seconds, and rendering in tens of milliseconds at a resolution of \({1920!\times!1080}\).

As research interest surges, vector similarity search is applied in multiple fields, including data mining, computer vision, and information retrieval. {Given a set of objects (e.g., a set of images) and a query object, we can easily transform each object into a feature vector and apply the vector similarity search to retrieve the most similar objects. However, the original vector similarity search cannot well support \textit{hybrid queries}, where users not only input unstructured query constraint (i.e., the feature vector of query object) but also structured query constraint (i.e., the desired attributes of interest). Hybrid query processing aims at identifying these objects with similar feature vectors to query object and satisfying the given attribute constraints. Recent efforts have attempted to answer a hybrid query by performing attribute filtering and vector similarity search separately and then merging the results later, which limits efficiency and accuracy because they are not purpose-built for hybrid queries.} In this paper, we propose a native hybrid query (NHQ) framework based on proximity graph (PG), which provides the specialized \textit{composite index and joint pruning} modules for hybrid queries. We easily deploy existing various PGs on this framework to process hybrid queries efficiently. Moreover, we present two novel navigable PGs (NPGs) with optimized edge selection and routing strategies, which obtain better overall performance than existing PGs. After that, we deploy the proposed NPGs in NHQ to form two hybrid query methods, which significantly outperform the state-of-the-art competitors on all experimental datasets (10\(\times\) faster under the same \textit{Recall}), including eight public and one in-house real-world datasets. Our code and datasets have been released at \url{https://github.com/AshenOn3/NHQ}.

The distributed representation of symbols is one of the key technologies in machine learning systems today, playing a pivotal role in modern natural language processing. Traditional word embeddings associate a separate vector with each word. While this approach is simple and leads to good performance, it requires a lot of memory for representing a large vocabulary. To reduce the memory footprint, the default embedding layer in spaCy is a hash embeddings layer. It is a stochastic approximation of traditional embeddings that provides unique vectors for a large number of words without explicitly storing a separate vector for each of them. To be able to compute meaningful representations for both known and unknown words, hash embeddings represent each word as a summary of the normalized word form, subword information and word shape. Together, these features produce a multi-embedding of a word. In this technical report we lay out a bit of history and introduce the embedding methods in spaCy in detail. Second, we critically evaluate the hash embedding architecture with multi-embeddings on Named Entity Recognition datasets from a variety of domains and languages. The experiments validate most key design choices behind spaCy’s embedders, but we also uncover a few surprising results.

Due to the availability of large-scale multi-modal data (e.g., satellite images acquired by different sensors, text sentences, etc) archives, the development of cross-modal retrieval systems that can search and retrieve semantically relevant data across different modalities based on a query in any modality has attracted great attention in RS. In this paper, we focus our attention on cross-modal text-image retrieval, where queries from one modality (e.g., text) can be matched to archive entries from another (e.g., image). Most of the existing cross-modal text-image retrieval systems require a high number of labeled training samples and also do not allow fast and memory-efficient retrieval due to their intrinsic characteristics. These issues limit the applicability of the existing cross-modal retrieval systems for large-scale applications in RS. To address this problem, in this paper we introduce a novel deep unsupervised cross-modal contrastive hashing (DUCH) method for RS text-image retrieval. The proposed DUCH is made up of two main modules: 1) feature extraction module (which extracts deep representations of the text-image modalities); and 2) hashing module (which learns to generate cross-modal binary hash codes from the extracted representations). Within the hashing module, we introduce a novel multi-objective loss function including: i) contrastive objectives that enable similarity preservation in both intra- and inter-modal similarities; ii) an adversarial objective that is enforced across two modalities for cross-modal representation consistency; iii) binarization objectives for generating representative hash codes. Experimental results show that the proposed DUCH outperforms state-of-the-art unsupervised cross-modal hashing methods on two multi-modal (image and text) benchmark archives in RS. Our code is publicly available at https://git.tu-berlin.de/rsim/duch.

Training of Machine Learning (ML) models in real contexts often deals with big data sets and high-class imbalance samples where the class of interest is unrepresented (minority class). Practical solutions using classical ML models address the problem of large data sets using parallel/distributed implementations of training algorithms, approximate model-based solutions, or applying instance selection (IS) algorithms to eliminate redundant information. However, the combined problem of big and high imbalanced datasets has been less addressed. This work proposes three new methods for IS to be able to deal with large and imbalanced data sets. The proposed methods use Locality Sensitive Hashing (LSH) as a base clustering technique, and then three different sampling methods are applied on top of the clusters (or buckets) generated by LSH. The algorithms were developed in the Apache Spark framework, guaranteeing their scalability. The experiments carried out in three different datasets suggest that the proposed IS methods can improve the performance of a base ML model between 5% and 19% in terms of the geometric mean.

Content-based file matching has been widely deployed for decades, largely for the detection of sources of copyright infringement, extremist materials, and abusive sexual media. Perceptual hashes, such as Microsoft’s PhotoDNA, are one automated mechanism for facilitating detection, allowing for machines to approximately match visual features of an image or video in a robust manner. However, there does not appear to be much public evaluation of such approaches, particularly when it comes to how effective they are against content-preserving modifications to media files. In this paper, we present a million-image scale evaluation of several perceptual hashing archetypes for popular algorithms (including Facebook’s PDQ, Apple’s Neuralhash, and the popular pHash library) against seven image variants. The focal point is the distribution of Hamming distance scores between both unrelated images and image variants to better understand the problems faced by each approach.

Fast approximations to matrix multiplication have the potential to dramatically reduce the cost of neural network inference. Recent work on approximate matrix multiplication proposed to replace costly multiplications with table-lookups by fitting a fast hash function from training data. In this work, we propose improvements to this previous work, targeted to the deep learning inference setting, where one has access to both training data and fixed (already learned) model weight matrices. We further propose a fine-tuning procedure for accelerating entire neural networks while minimizing loss in accuracy. Finally, we analyze the proposed method on a simple image classification task. While we show improvements to prior work, overall classification accuracy remains substantially diminished compared to exact matrix multiplication. Our work, despite this negative result, points the way towards future efforts to accelerate inner products with fast nonlinear hashing methods.

The development of cross-modal retrieval systems that can search and retrieve semantically relevant data across different modalities based on a query in any modality has attracted great attention in remote sensing (RS). In this paper, we focus our attention on cross-modal text-image retrieval, where queries from one modality (e.g., text) can be matched to archive entries from another (e.g., image). Most of the existing cross-modal text-image retrieval systems in RS require a high number of labeled training samples and also do not allow fast and memory-efficient retrieval. These issues limit the applicability of the existing cross-modal retrieval systems for large-scale applications in RS. To address this problem, in this paper we introduce a novel unsupervised cross-modal contrastive hashing (DUCH) method for text-image retrieval in RS. To this end, the proposed DUCH is made up of two main modules: 1) feature extraction module, which extracts deep representations of two modalities; 2) hashing module that learns to generate cross-modal binary hash codes from the extracted representations. We introduce a novel multi-objective loss function including: i) contrastive objectives that enable similarity preservation in intra- and inter-modal similarities; ii) an adversarial objective that is enforced across two modalities for cross-modal representation consistency; and iii) binarization objectives for generating hash codes. Experimental results show that the proposed DUCH outperforms state-of-the-art methods. Our code is publicly available at https://git.tu-berlin.de/rsim/duch.

We present HashEncoding, a novel autoencoding architecture that leverages a non-parametric multiscale coordinate hash function to facilitate a per-pixel decoder without convolutions. By leveraging the space-folding behaviour of hashing functions, HashEncoding allows for an inherently multiscale embedding space that remains much smaller than the original image. As a result, the decoder requires very few parameters compared with decoders in traditional autoencoders, approaching a non-parametric reconstruction of the original image and allowing for greater generalizability. Finally, by allowing backpropagation directly to the coordinate space, we show that HashEncoding can be exploited for geometric tasks such as optical flow.

Binary codes are constructed from incidence matrices of hypergraphs. A combinatroial description is given for the minimum distances of such codes via a combinatorial tool called ``eonv”. This combinatorial approach provides a faster alternative method of finding the minimum distance, which is known to be a hard problem. This is demonstrated on several classes of codes from hypergraphs. Moreover, self-duality and self-orthogonality conditions are also studied through hypergraphs.

We present a deep learning approach for learning the joint semantic embeddings of images and captions in a Euclidean space, such that the semantic similarity is approximated by the L2 distances in the embedding space. For that, we introduce a metric learning scheme that utilizes multitask learning to learn the embedding of identical semantic concepts using a center loss. By introducing a differentiable quantization scheme into the end-to-end trainable network, we derive a semantic embedding of semantically similar concepts in Euclidean space. We also propose a novel metric learning formulation using an adaptive margin hinge loss, that is refined during the training phase. The proposed scheme was applied to the MS-COCO, Flicke30K and Flickr8K datasets, and was shown to compare favorably with contemporary state-of-the-art approaches.

On 2\(^{nd}\) October 2012 the NIST (National Institute of Standards and Technology) in the United States of America announced the new hashing algorithm which will be adopted as standard from now on. Among a total of 73 candidates, the winner was Keccak, designed by a group of cryptographers from Belgium and Italy. The public selection of a new standard of cryptographic hash function SHA (Secure Hash Algorithm) took five years. Its object is to generate a hash a fixed size from a pattern with arbitrary length. The first selection on behalf of NIST on a standard of this family took place in 1993 when SHA-1 was chosen, which later on was replaced by SHA-2. This paper is focused on the analysis both from the point of view of security and the implementation of the Keccak function, which is the base of the new SHA-3 standard. In particular, an implementation in the mobile platform Android is presented here, providing the first known external library in this mobile operating system so that any developer could use the new standard hashing. Finally, the new standard in applications in the Internet of Things is analysed.

As a novel deep learning model, gcForest has been widely used in various applications. However, the current multi-grained scanning of gcForest produces many redundant feature vectors, and this increases the time cost of the model. To screen out redundant feature vectors, we introduce a hashing screening mechanism for multi-grained scanning and propose a model called HW-Forest which adopts two strategies, hashing screening and window screening. HW-Forest employs perceptual hashing algorithm to calculate the similarity between feature vectors in hashing screening strategy, which is used to remove the redundant feature vectors produced by multi-grained scanning and can significantly decrease the time cost and memory consumption. Furthermore, we adopt a self-adaptive instance screening strategy to improve the performance of our approach, called window screening, which can achieve higher accuracy without hyperparameter tuning on different datasets. Our experimental results show that HW-Forest has higher accuracy than other models, and the time cost is also reduced.

Thanks to the efficient retrieval speed and low storage consumption, learning to hash has been widely used in visual retrieval tasks. However, existing hashing methods assume that the query and retrieval samples lie in homogeneous feature space within the same domain. As a result, they cannot be directly applied to heterogeneous cross-domain retrieval. In this paper, we propose a Generalized Image Transfer Retrieval (GITR) problem, which encounters two crucial bottlenecks: 1) the query and retrieval samples may come from different domains, leading to an inevitable {domain distribution gap}; 2) the features of the two domains may be heterogeneous or misaligned, bringing up an additional {feature gap}. To address the GITR problem, we propose an Asymmetric Transfer Hashing (ATH) framework with its unsupervised/semi-supervised/supervised realizations. Specifically, ATH characterizes the domain distribution gap by the discrepancy between two asymmetric hash functions, and minimizes the feature gap with the help of a novel adaptive bipartite graph constructed on cross-domain data. By jointly optimizing asymmetric hash functions and the bipartite graph, not only can knowledge transfer be achieved but information loss caused by feature alignment can also be avoided. Meanwhile, to alleviate negative transfer, the intrinsic geometrical structure of single-domain data is preserved by involving a domain affinity graph. Extensive experiments on both single-domain and cross-domain benchmarks under different GITR subtasks indicate the superiority of our ATH method in comparison with the state-of-the-art hashing methods.

Hierarchical semantic structures, naturally existing in real-world datasets, can assist in capturing the latent distribution of data to learn robust hash codes for retrieval systems. Although hierarchical semantic structures can be simply expressed by integrating semantically relevant data into a high-level taxon with coarser-grained semantics, the construction, embedding, and exploitation of the structures remain tricky for unsupervised hash learning. To tackle these problems, we propose a novel unsupervised hashing method named Hyperbolic Hierarchical Contrastive Hashing (HHCH). We propose to embed continuous hash codes into hyperbolic space for accurate semantic expression since embedding hierarchies in hyperbolic space generates less distortion than in hyper-sphere space and Euclidean space. In addition, we extend the K-Means algorithm to hyperbolic space and perform the proposed hierarchical hyperbolic K-Means algorithm to construct hierarchical semantic structures adaptively. To exploit the hierarchical semantic structures in hyperbolic space, we designed the hierarchical contrastive learning algorithm, including hierarchical instance-wise and hierarchical prototype-wise contrastive learning. Extensive experiments on four benchmark datasets demonstrate that the proposed method outperforms the state-of-the-art unsupervised hashing methods. Codes will be released.

With the rapid growth of multimodal media data on the Web in recent years, hash learning methods as a way to achieve efficient and flexible cross-modal retrieval of massive multimedia data have received a lot of attention from the current Web resource retrieval research community. Existing supervised hashing methods simply transform label information into pairwise similarity information to guide hash learning, leading to a potential risk of semantic error in the face of multi-label data. In addition, most existing hash optimization methods solve NP-hard optimization problems by employing approximate approximation strategies based on relaxation strategies, leading to a large quantization error. In order to address above obstacles, we present a simple yet efficient Adaptive Asymmetric Label-guided Hashing, named A2LH, for Multimedia Search. Specifically, A2LH is a two-step hashing method. In the first step, we design an association representation model between the different modality representations and semantic label representation separately, and use the semantic label representation as an intermediate bridge to solve the semantic gap existing between different modalities. In addition, we present an efficient discrete optimization algorithm for solving the quantization error problem caused by relaxation-based optimization algorithms. In the second step, we leverage the generated hash codes to learn the hash mapping functions. The experimental results show that our proposed method achieves optimal performance on all compared baseline methods.

Existing unsupervised hash learning is a kind of attribute-centered calculation. It may not accurately preserve the similarity between data. This leads to low down the performance of hash function learning. In this paper, a hash algorithm is proposed with a hyper-class representation. It is a two-steps approach. The first step finds potential decision features and establish hyper-class. The second step constructs hash learning based on the hyper-class information in the first step, so that the hash codes of the data within the hyper-class are as similar as possible, as well as the hash codes of the data between the hyper-classes are as different as possible. To evaluate the efficiency, a series of experiments are conducted on four public datasets. The experimental results show that the proposed hash algorithm is more efficient than the compared algorithms, in terms of mean average precision (MAP), average precision (AP) and Hamming radius 2 (HAM2)

Recently, deep cross-modal hashing has gained increasing attention. However, in many practical cases, data are distributed and cannot be collected due to privacy concerns, which greatly reduces the cross-modal hashing performance on each client. And due to the problems of statistical heterogeneity, model heterogeneity, and forcing each client to accept the same parameters, applying federated learning to cross-modal hash learning becomes very tricky. In this paper, we propose a novel method called prototype-based layered federated cross-modal hashing. Specifically, the prototype is introduced to learn the similarity between instances and classes on server, reducing the impact of statistical heterogeneity (non-IID) on different clients. And we monitor the distance between local and global prototypes to further improve the performance. To realize personalized federated learning, a hypernetwork is deployed on server to dynamically update different layers’ weights of local model. Experimental results on benchmark datasets show that our method outperforms state-of-the-art methods.

Deep unsupervised hashing has been appreciated in the regime of image retrieval. However, most prior arts failed to detect the semantic components and their relationships behind the images, which makes them lack discriminative power. To make up the defect, we propose a novel Deep Semantic Components Hashing (DSCH), which involves a common sense that an image normally contains a bunch of semantic components with homology and co-occurrence relationships. Based on this prior, DSCH regards the semantic components as latent variables under the Expectation-Maximization framework and designs a two-step iterative algorithm with the objective of maximum likelihood of training data. Firstly, DSCH constructs a semantic component structure by uncovering the fine-grained semantics components of images with a Gaussian Mixture Modal~(GMM), where an image is represented as a mixture of multiple components, and the semantics co-occurrence are exploited. Besides, coarse-grained semantics components, are discovered by considering the homology relationships between fine-grained components, and the hierarchy organization is then constructed. Secondly, DSCH makes the images close to their semantic component centers at both fine-grained and coarse-grained levels, and also makes the images share similar semantic components close to each other. Extensive experiments on three benchmark datasets demonstrate that the proposed hierarchical semantic components indeed facilitate the hashing model to achieve superior performance.

Cross-modal hashing is an important approach for multimodal data management and application. Existing unsupervised cross-modal hashing algorithms mainly rely on data features in pre-trained models to mine their similarity relationships. However, their optimization objectives are based on the static metric between the original uni-modal features, without further exploring data correlations during the training. In addition, most of them mainly focus on association mining and alignment among pairwise instances in continuous space but ignore the latent structural correlations contained in the semantic hashing space. In this paper, we propose an unsupervised hash learning framework, namely Adaptive Structural Similarity Preservation Hashing (ASSPH), to solve the above problems. Firstly, we propose an adaptive learning scheme, with limited data and training batches, to enrich semantic correlations of unlabeled instances during the training process and meanwhile to ensure a smooth convergence of the training process. Secondly, we present an asymmetric structural semantic representation learning scheme. We introduce structural semantic metrics based on graph adjacency relations during the semantic reconstruction and correlation mining stage and meanwhile align the structure semantics in the hash space with an asymmetric binary optimization process. Finally, we conduct extensive experiments to validate the enhancements of our work in comparison with existing works.

Cross-modal hashing is a successful method to solve large-scale multimedia retrieval issue. A lot of matrix factorization-based hashing methods are proposed. However, the existing methods still struggle with a few problems, such as how to generate the binary codes efficiently rather than directly relax them to continuity. In addition, most of the existing methods choose to use an \(n\times n\) similarity matrix for optimization, which makes the memory and computation unaffordable. In this paper we propose a novel Asymmetric Scalable Cross-Modal Hashing (ASCMH) to address these issues. It firstly introduces a collective matrix factorization to learn a common latent space from the kernelized features of different modalities, and then transforms the similarity matrix optimization to a distance-distance difference problem minimization with the help of semantic labels and common latent space. Hence, the computational complexity of the \(n\times n\) asymmetric optimization is relieved. In the generation of hash codes we also employ an orthogonal constraint of label information, which is indispensable for search accuracy. So the redundancy of computation can be much reduced. For efficient optimization and scalable to large-scale datasets, we adopt the two-step approach rather than optimizing simultaneously. Extensive experiments on three benchmark datasets: Wiki, MIRFlickr-25K, and NUS-WIDE, demonstrate that our ASCMH outperforms the state-of-the-art cross-modal hashing methods in terms of accuracy and efficiency.

In collaborative outlier detection, multiple participants exchange their local detectors trained on decentralized devices without exchanging their own data. A key problem of collaborative outlier detection is efficiently aggregating multiple local detectors to form a global detector without breaching the privacy of participants’ data and degrading the detection accuracy. We study locality-sensitive hashing-based ensemble methods to detect collaborative outliers since they are mergeable and compatible with differentially private mechanisms. Our proposed LSH iTables is simple and outperforms recent ensemble competitors on centralized and decentralized scenarios over many real-world data sets.

Remote sensing image retrieval (RSIR), aiming at searching for a set of similar items to a given query image, is a very important task in remote sensing applications. Deep hashing learning as the current mainstream method has achieved satisfactory retrieval performance. On one hand, various deep neural networks are used to extract semantic features of remote sensing images. On the other hand, the hashing techniques are subsequently adopted to map the high-dimensional deep features to the low-dimensional binary codes. This kind of methods attempts to learn one hash function for both the query and database samples in a symmetric way. However, with the number of database samples increasing, it is typically time-consuming to generate the hash codes of large-scale database images. In this paper, we propose a novel deep hashing method, named asymmetric hash code learning (AHCL), for RSIR. The proposed AHCL generates the hash codes of query and database images in an asymmetric way. In more detail, the hash codes of query images are obtained by binarizing the output of the network, while the hash codes of database images are directly learned by solving the designed objective function. In addition, we combine the semantic information of each image and the similarity information of pairs of images as supervised information to train a deep hashing network, which improves the representation ability of deep features and hash codes. The experimental results on three public datasets demonstrate that the proposed method outperforms symmetric methods in terms of retrieval accuracy and efficiency. The source code is available at https://github.com/weiweisong415/Demo AHCL for TGRS2022.

Analyzing numerous or long time series is difficult in practice due to the high storage costs and computational requirements. Therefore, techniques have been proposed to generate compact similarity-preserving representations of time series, enabling real-time similarity search on large in-memory data collections. However, the existing techniques are not ideally suited for assessing similarity when sequences are locally out of phase. In this paper, we propose the use of product quantization for efficient similarity-based comparison of time series under time warping. The idea is to first compress the data by partitioning the time series into equal length sub-sequences which are represented by a short code. The distance between two time series can then be efficiently approximated by pre-computed elastic distances between their codes. The partitioning into sub-sequences forces unwanted alignments, which we address with a pre-alignment step using the maximal overlap discrete wavelet transform (MODWT). To demonstrate the efficiency and accuracy of our method, we perform an extensive experimental evaluation on benchmark datasets in nearest neighbors classification and clustering applications. Overall, the proposed solution emerges as a highly efficient (both in terms of memory usage and computation time) replacement for elastic measures in time series applications.

Existing data-dependent hashing methods use large backbone networks with millions of parameters and are computationally complex. Existing knowledge distillation methods use logits and other features of the deep (teacher) model and as knowledge for the compact (student) model, which requires the teacher’s network to be fine-tuned on the context in parallel with the student model on the context. Training teacher on the target context requires more time and computational resources. In this paper, we propose context unaware knowledge distillation that uses the knowledge of the teacher model without fine-tuning it on the target context. We also propose a new efficient student model architecture for knowledge distillation. The proposed approach follows a two-step process. The first step involves pre-training the student model with the help of context unaware knowledge distillation from the teacher model. The second step involves fine-tuning the student model on the context of image retrieval. In order to show the efficacy of the proposed approach, we compare the retrieval results, no. of parameters and no. of operations of the student models with the teacher models under different retrieval frameworks, including deep cauchy hashing (DCH) and central similarity quantization (CSQ). The experimental results confirm that the proposed approach provides a promising trade-off between the retrieval results and efficiency. The code used in this paper is released publicly at \url{https://github.com/satoru2001/CUKDFIR}.

Efficient document retrieval heavily relies on the technique of semantic hashing, which learns a binary code for every document and employs Hamming distance to evaluate document distances. However, existing semantic hashing methods are mostly established on outdated TFIDF features, which obviously do not contain lots of important semantic information about documents. Furthermore, the Hamming distance can only be equal to one of several integer values, significantly limiting its representational ability for document distances. To address these issues, in this paper, we propose to leverage BERT embeddings to perform efficient retrieval based on the product quantization technique, which will assign for every document a real-valued codeword from the codebook, instead of a binary code as in semantic hashing. Specifically, we first transform the original BERT embeddings via a learnable mapping and feed the transformed embedding into a probabilistic product quantization module to output the assigned codeword. The refining and quantizing modules can be optimized in an end-to-end manner by minimizing the probabilistic contrastive loss. A mutual information maximization based method is further proposed to improve the representativeness of codewords, so that documents can be quantized more accurately. Extensive experiments conducted on three benchmarks demonstrate that our proposed method significantly outperforms current state-of-the-art baselines.

In this paper, we propose Adam-Hash: an adaptive and dynamic multi-resolution hashing data-structure for fast pairwise summation estimation. Given a data-set \(X \subset \mathbb{R}^d\), a binary function \(f:\mathbb{R}^d\times \mathbb{R}^d\to \mathbb{R}\), and a point \(y \in \mathbb{R}^d\), the Pairwise Summation Estimate \(\mathrm{PSE}X(y) := \frac{1}{|X|} \sum{x \in X} f(x,y)\). For any given data-set \(X\), we need to design a data-structure such that given any query point \(y \in \mathbb{R}^d\), the data-structure approximately estimates \(\mathrm{PSE}X(y)\) in time that is sub-linear in \(|X|\). Prior works on this problem have focused exclusively on the case where the data-set is static, and the queries are independent. In this paper, we design a hashing-based PSE data-structure which works for the more practical \textit{dynamic} setting in which insertions, deletions, and replacements of points are allowed. Moreover, our proposed Adam-Hash is also robust to adaptive PSE queries, where an adversary can choose query \(q_j \in \mathbb{R}^d\) depending on the output from previous queries \(q_1, q_2, \dots, q{j-1}\).

Vertical Federated Learning (VFL) is a trending collaborative machine learning model training solution. Existing industrial frameworks employ secure multi-party computation techniques such as homomorphic encryption to ensure data security and privacy. Despite these efforts, studies have revealed that data leakage remains a risk in VFL due to the correlations between intermediate representations and raw data. Neural networks can accurately capture these correlations, allowing an adversary to reconstruct the data. This emphasizes the need for continued research into securing VFL systems. Our work shows that hashing is a promising solution to counter data reconstruction attacks. The one-way nature of hashing makes it difficult for an adversary to recover data from hash codes. However, implementing hashing in VFL presents new challenges, including vanishing gradients and information loss. To address these issues, we propose HashVFL, which integrates hashing and simultaneously achieves learnability, bit balance, and consistency. Experimental results indicate that HashVFL effectively maintains task performance while defending against data reconstruction attacks. It also brings additional benefits in reducing the degree of label leakage, mitigating adversarial attacks, and detecting abnormal inputs. We hope our work will inspire further research into the potential applications of HashVFL.

Minimal perfect hashing is the problem of mapping a static set of \(n\) distinct keys into the address space \(\{1,\ldots,n\}\) bijectively. It is well-known that \(nlog_2(e)\) bits are necessary to specify a minimal perfect hash function (MPHF) \(f\), when no additional knowledge of the input keys is to be used. However, it is often the case in practice that the input keys have intrinsic relationships that we can exploit to lower the bit complexity of \(f\). For example, consider a string and the set of all its distinct \(k\)-mers as input keys: since two consecutive \(k\)-mers share an overlap of \(k-1\) symbols, it seems possible to beat the classic \(log_2(e)\) bits/key barrier in this case. Moreover, we would like \(f\) to map consecutive \(k\)-mers to consecutive addresses, as to also preserve as much as possible their relationship in the codomain. This is a useful feature in practice as it guarantees a certain degree of locality of reference for \(f\), resulting in a better evaluation time when querying consecutive \(k\)-mers. Motivated by these premises, we initiate the study of a new type of locality-preserving MPHF designed for \(k\)-mers extracted consecutively from a collection of strings. We design a construction whose space usage decreases for growing \(k\) and discuss experiments with a practical implementation of the method: in practice, the functions built with our method can be several times smaller and even faster to query than the most efficient MPHFs in the literature.

We present Falconn++, a novel locality-sensitive filtering (LSF) approach for approximate nearest neighbor search on angular distance. Falconn++ can filter out potential far away points in any hash bucket before querying, which results in higher quality candidates compared to other hashing-based solutions. Theoretically, Falconn++ asymptotically achieves lower query time complexity than Falconn, an optimal locality-sensitive hashing scheme on angular distance. Empirically, Falconn++ achieves a higher recall-speed tradeoff than Falconn on many real-world data sets. Falconn++ is also competitive with HNSW, an efficient representative of graph-based solutions on high search recall regimes.

In deep learning, embeddings are widely used to represent categorical entities such as words, apps, and movies. An embedding layer maps each entity to a unique vector, causing the layer’s memory requirement to be proportional to the number of entities. In the recommendation domain, a given category can have hundreds of thousands of entities, and its embedding layer can take gigabytes of memory. The scale of these networks makes them difficult to deploy in resource constrained environments. In this paper, we propose a novel approach for reducing the size of an embedding table while still mapping each entity to its own unique embedding. Rather than maintaining the full embedding table, we construct each entity’s embedding “on the fly” using two separate embedding tables. The first table employs hashing to force multiple entities to share an embedding. The second table contains one trainable weight per entity, allowing the model to distinguish between entities sharing the same embedding. Since these two tables are trained jointly, the network is able to learn a unique embedding per entity, helping it maintain a discriminative capability similar to a model with an uncompressed embedding table. We call this approach MEmCom (Multi-Embedding Compression). We compare with state-of-the-art model compression techniques for multiple problem classes including classification and ranking. On four popular recommender system datasets, MEmCom had a 4% relative loss in nDCG while compressing the input embedding sizes of our recommendation models by 16x, 4x, 12x, and 40x. MEmCom outperforms the state-of-the-art techniques, which achieved 16%, 6%, 10%, and 8% relative loss in nDCG at the respective compression ratios. Additionally, MEmCom is able to compress the RankNet ranking model by 32x on a dataset with millions of users’ interactions with games while incurring only a 1% relative loss in nDCG.

Modern hash table designs strive to minimize space while maximizing speed. The most important factor in speed is the number of cache lines accessed during updates and queries. This is especially important on PMEM, which is slower than DRAM and in which writes are more expensive than reads. This paper proposes two stronger design objectives: stability and low-associativity. A stable hash table doesn’t move items around, and a hash table has low associativity if there are only a few locations where an item can be stored. Low associativity ensures that queries need to examine only a few memory locations, and stability ensures that insertions write to very few cache lines. Stability also simplifies scaling and crash safety. We present IcebergHT, a fast, crash-safe, concurrent, and space-efficient hash table for PMEM based on the design principles of stability and low associativity. IcebergHT combines in-memory metadata with a new hashing technique, iceberg hashing, that is (1) space efficient, (2) stable, and (3) supports low associativity. In contrast, existing hash-tables either modify numerous cache lines during insertions (e.g. cuckoo hashing), access numerous cache lines during queries (e.g. linear probing), or waste space (e.g. chaining). Moreover, the combination of (1)-(3) yields several emergent benefits: IcebergHT scales better than other hash tables, supports crash-safety, and has excellent performance on PMEM (where writes are particularly expensive).

Image forensics is a rising topic as the trustworthy multimedia content is critical for modern society. Like other vision-related applications, forensic analysis relies heavily on the proper image representation. Despite the importance, current theoretical understanding for such representation remains limited, with varying degrees of neglect for its key role. For this gap, we attempt to investigate the forensic-oriented image representation as a distinct problem, from the perspectives of theory, implementation, and application. Our work starts from the abstraction of basic principles that the representation for forensics should satisfy, especially revealing the criticality of robustness, interpretability, and coverage. At the theoretical level, we propose a new representation framework for forensics, called Dense Invariant Representation (DIR), which is characterized by stable description with mathematical guarantees. At the implementation level, the discrete calculation problems of DIR are discussed, and the corresponding accurate and fast solutions are designed with generic nature and constant complexity. We demonstrate the above arguments on the dense-domain pattern detection and matching experiments, providing comparison results with state-of-the-art descriptors. Also, at the application level, the proposed DIR is initially explored in passive and active forensics, namely copy-move forgery detection and perceptual hashing, exhibiting the benefits in fulfilling the requirements of such forensic tasks.

Zero-shot cross-modal retrieval (ZS-CMR) deals with the retrieval problem among heterogenous data from unseen classes. Typically, to guarantee generalization, the pre-defined class embeddings from natural language processing (NLP) models are used to build a common space. In this paper, instead of using an extra NLP model to define a common space beforehand, we consider a totally different way to construct (or learn) a common hamming space from an information-theoretic perspective. We term our model the Information-Theoretic Hashing (ITH), which is composed of two cascading modules: an Adaptive Information Aggregation (AIA) module; and a Semantic Preserving Encoding (SPE) module. Specifically, our AIA module takes the inspiration from the Principle of Relevant Information (PRI) to construct a common space that adaptively aggregates the intrinsic semantics of different modalities of data and filters out redundant or irrelevant information. On the other hand, our SPE module further generates the hashing codes of different modalities by preserving the similarity of intrinsic semantics with the element-wise Kullback-Leibler (KL) divergence. A total correlation regularization term is also imposed to reduce the redundancy amongst different dimensions of hash codes. Sufficient experiments on three benchmark datasets demonstrate the superiority of the proposed ITH in ZS-CMR. Source code is available in the supplementary material.

Cross-modal retrieval aims to search for data with similar semantic meanings across different content modalities. However, cross-modal retrieval requires huge amounts of storage and retrieval time since it needs to process data in multiple modalities. Existing works focused on learning single-source compact features such as binary hash codes that preserve similarities between different modalities. In this work, we propose a jointly learned deep hashing and quantization network (HQ) for cross-modal retrieval. We simultaneously learn binary hash codes and quantization codes to preserve semantic information in multiple modalities by an end-to-end deep learning architecture. At the retrieval step, binary hashing is used to retrieve a subset of items from the search space, then quantization is used to re-rank the retrieved items. We theoretically and empirically show that this two-stage retrieval approach provides faster retrieval results while preserving accuracy. Experimental results on the NUS-WIDE, MIR-Flickr, and Amazon datasets demonstrate that HQ achieves boosts of more than 7% in precision compared to supervised neural network-based compact coding models.

Hashing-based Recommender Systems (RSs) are widely studied to provide scalable services. The existing methods for the systems combine three modules to achieve efficiency: feature extraction, interaction modeling, and binarization. In this paper, we study an unexplored module combination for the hashing-based recommender systems, namely Compact Cross-Similarity Recommender (CCSR). Inspired by cross-modal retrieval, CCSR utilizes Maximum a Posteriori similarity instead of matrix factorization and rating reconstruction to model interactions between users and items. We conducted experiments on MovieLens1M, Amazon product review, Ichiba purchase dataset and confirmed CCSR outperformed the existing matrix factorization-based methods. On the Movielens1M dataset, the absolute performance improvements are up to 15.69% in NDCG and 4.29% in Recall. In addition, we extensively studied three binarization modules: \(sign\), scaled tanh, and sign-scaled tanh. The result demonstrated that although differentiable scaled tanh is popular in recent discrete feature learning literature, a huge performance drop occurs when outputs of scaled \(tanh\) are forced to be binary.

Hashing that projects data into binary codes has shown extraordinary talents in cross-modal retrieval due to its low storage usage and high query speed. Despite their empirical success on some scenarios, existing cross-modal hashing methods usually fail to cross modality gap when fully-paired data with plenty of labeled information is nonexistent. To circumvent this drawback, motivated by the Divide-and-Conquer strategy, we propose Deep Manifold Hashing (DMH), a novel method of dividing the problem of semi-paired unsupervised cross-modal retrieval into three sub-problems and building one simple yet efficiency model for each sub-problem. Specifically, the first model is constructed for obtaining modality-invariant features by complementing semi-paired data based on manifold learning, whereas the second model and the third model aim to learn hash codes and hash functions respectively. Extensive experiments on three benchmarks demonstrate the superiority of our DMH compared with the state-of-the-art fully-paired and semi-paired unsupervised cross-modal hashing methods.

In this paper, we propose Suppression-Enhancing Mask based attention and Interactive Channel transformatiON (SEMICON) to learn binary hash codes for dealing with large-scale fine-grained image retrieval tasks. In SEMICON, we first develop a suppression-enhancing mask (SEM) based attention to dynamically localize discriminative image regions. More importantly, different from existing attention mechanism simply erasing previous discriminative regions, our SEM is developed to restrain such regions and then discover other complementary regions by considering the relation between activated regions in a stage-by-stage fashion. In each stage, the interactive channel transformation (ICON) module is afterwards designed to exploit correlations across channels of attended activation tensors. Since channels could generally correspond to the parts of fine-grained objects, the part correlation can be also modeled accordingly, which further improves fine-grained retrieval accuracy. Moreover, to be computational economy, ICON is realized by an efficient two-step process. Finally, the hash learning of our SEMICON consists of both global- and local-level branches for better representing fine-grained objects and then generating binary hash codes explicitly corresponding to multiple levels. Experiments on five benchmark fine-grained datasets show our superiority over competing methods.

This paper investigates theoretical properties of subsampling and hashing as tools for approximate Euclidean norm-preserving embeddings for vectors with (unknown) additive Gaussian noises. Such embeddings are sometimes called Johnson-lindenstrauss embeddings due to their celebrated lemma. Previous work shows that as sparse embeddings, the success of subsampling and hashing closely depends on the \(l_\infty\) to \(l_2\) ratios of the vector to be mapped. This paper shows that the presence of noise removes such constrain in high-dimensions, in other words, sparse embeddings such as subsampling and hashing with comparable embedding dimensions to dense embeddings have similar approximate norm-preserving dimensionality-reduction properties. The key is that the noise should be treated as an information to be exploited, not simply something to be removed. Theoretical bounds for subsampling and hashing to recover the approximate norm of a high dimension vector in the presence of noise are derived, with numerical illustrations showing better performances are achieved in the presence of noise.

Applications of face recognition systems for authentication purposes are growing rapidly. Although state-of-the-art (SOTA) face recognition systems have high recognition accuracy, the features which are extracted for each user and are stored in the system’s database contain privacy-sensitive information. Accordingly, compromising this data would jeopardize users’ privacy. In this paper, we propose a new cancelable template protection method, dubbed MLP-hash, which generates protected templates by passing the extracted features through a user-specific randomly-weighted multi-layer perceptron (MLP) and binarizing the MLP output. We evaluated the unlinkability, irreversibility, and recognition accuracy of our proposed biometric template protection method to fulfill the ISO/IEC 30136 standard requirements. Our experiments with SOTA face recognition systems on the MOBIO and LFW datasets show that our method has competitive performance with the BioHashing and IoM Hashing (IoM-GRP and IoM-URP) template protection algorithms. We provide an open-source implementation of all the experiments presented in this paper so that other researchers can verify our findings and build upon our work.

Frequency estimation of elements is an important task for summarizing data streams and machine learning applications. The problem is often addressed by using streaming algorithms with sublinear space data structures. These algorithms allow processing of large data while using limited data storage. Commonly used streaming algorithms, such as count-min sketch, have many advantages, but do not take into account properties of a data stream for performance optimization. In the present paper we introduce a novel double-hashing algorithm that provides flexibility to optimize streaming algorithms depending on the properties of a given stream. In the double-hashing approach, first a standard streaming algorithm is employed to obtain an estimate of the element frequencies. This estimate is derived using a fraction of the stream and allows identification of the heavy hitters. Next, it uses a modified hash table where the heavy hitters are mapped into individual buckets and other stream elements are mapped into the remaining buckets. Finally, the element frequencies are estimated based on the constructed hash table over the entire data stream with any streaming algorithm. We demonstrate on both synthetic data and an internet query log dataset that our approach is capable of improving frequency estimation due to removing heavy hitters from the hashing process and, thus, reducing collisions in the hash table. Our approach avoids employing additional machine learning models to identify heavy hitters and, thus, reduces algorithm complexity and streamlines implementation. Moreover, because it is not dependent on specific features of the stream elements for identifying heavy hitters, it is applicable to a large variety of streams. In addition, we propose a procedure on how to dynamically adjust the proposed double-hashing algorithm when frequencies of the elements in a stream are changing over time.

Modern high load applications store data using multiple database instances. Such an architecture requires data consistency, and it is important to ensure even distribution of data among nodes. Load balancing is used to achieve these goals. Hashing is the backbone of virtually all load balancing systems. Since the introduction of classic Consistent Hashing, many algorithms have been devised for this purpose. One of the purposes of the load balancer is to ensure storage cluster scalability. It is crucial for the performance of the whole system to transfer as few data records as possible during node addition or removal. The load balancer hashing algorithm has the greatest impact on this process. In this paper we experimentally evaluate several hashing algorithms used for load balancing, conducting both simulated and real system experiments. To evaluate algorithm performance, we have developed a benchmark suite based on Unidata MDM~ – a scalable toolkit for various Master Data Management (MDM) applications. For assessment, we have employed three criteria~ – uniformity of the produced distribution, the number of moved records, and computation speed. Following the results of our experiments, we have created a table, in which each algorithm is given an assessment according to the abovementioned criteria.

Identifying near duplicates within large, noisy text corpora has a myriad of applications that range from de-duplicating training datasets, reducing privacy risk, and evaluating test set leakage, to identifying reproduced news articles and literature within large corpora. Across these diverse applications, the overwhelming majority of work relies on N-grams. Limited efforts have been made to evaluate how well N-gram methods perform, in part because it is unclear how one could create an unbiased evaluation dataset for a massive corpus. This study uses the unique timeliness of historical news wires to create a 27,210 document dataset, with 122,876 positive duplicate pairs, for studying noise-robust de-duplication. The time-sensitivity of news makes comprehensive hand labelling feasible - despite the massive overall size of the corpus - as duplicates occur within a narrow date range. The study then develops and evaluates a range of de-duplication methods: hashing and N-gram overlap (which predominate in the literature), a contrastively trained bi-encoder, and a re-rank style approach combining a bi- and cross-encoder. The neural approaches significantly outperform hashing and N-gram overlap. We show that the bi-encoder scales well, de-duplicating a 10 million article corpus on a single GPU card in a matter of hours. We also apply our pre-trained model to the RealNews and patent portions of C4 (Colossal Clean Crawled Corpus), illustrating that a neural approach can identify many near duplicates missed by hashing, in the presence of various types of noise. The public release of our NEWS-COPY de-duplication dataset, codebase, and the pre-trained models will facilitate further research and applications.

We consider a similarity measure between two sets \(A\) and \(B\) of vectors, that balances the average and maximum cosine distance between pairs of vectors, one from set \(A\) and one from set \(B\). As a motivation for this measure, we present lineage tracking in a database. To practically realize this measure, we need an approximate search algorithm that given a set of vectors \(A\) and sets of vectors \(B_1,…,B_n\), the algorithm quickly locates the set \(B_i\) that maximizes the similarity measure. For the case where all sets are singleton sets, essentially each is a single vector, there are known efficient approximate search algorithms, e.g., approximated versions of tree search algorithms, locality-sensitive hashing (LSH), vector quantization (VQ) and proximity graph algorithms. In this work, we present approximate search algorithms for the general case. The underlying idea in these algorithms is encoding a set of vectors via a “long” single vector. The proposed approximate approach achieves significant performance gains over an optimized, exact search on vector sets.

The booming vector manage system calls for feasible similarity hash function as a front-end to perform similarity analysis. In this paper, we make a systematical survey on the existent well-known similarity hash functions to tease out the satisfied ones. We conclude that the similarity hash function MinHash and Nilsimsa can be directly marshaled into the pipeline of similarity analysis using vector manage system. After that, we make a brief and empirical discussion on the performance, drawbacks of the these functions and highlight MinHash, the variant of SimHash and feature hashing are the best for vector management system for large-scale similarity analysis.

Networks are ubiquitous in the real world. Link prediction, as one of the key problems for network-structured data, aims to predict whether there exists a link between two nodes. The traditional approaches are based on the explicit similarity computation between the compact node representation by embedding each node into a low-dimensional space. In order to efficiently handle the intensive similarity computation in link prediction, the hashing technique has been successfully used to produce the node representation in the Hamming space. However, the hashing-based link prediction algorithms face accuracy loss from the randomized hashing techniques or inefficiency from the learning to hash techniques in the embedding process. Currently, the Graph Neural Network (GNN) framework has been widely applied to the graph-related tasks in an end-to-end manner, but it commonly requires substantial computational resources and memory costs due to massive parameter learning, which makes the GNN-based algorithms impractical without the help of a powerful workhorse. In this paper, we propose a simple and effective model called #GNN, which balances the trade-off between accuracy and efficiency. #GNN is able to efficiently acquire node representation in the Hamming space for link prediction by exploiting the randomized hashing technique to implement message passing and capture high-order proximity in the GNN framework. Furthermore, we characterize the discriminative power of #GNN in probability. The extensive experimental results demonstrate that the proposed #GNN algorithm achieves accuracy comparable to the learning-based algorithms and outperforms the randomized algorithm, while running significantly faster than the learning-based algorithms. Also, the proposed algorithm shows excellent scalability on a large-scale network with the limited resources.

The mathematical formalization of a neurological mechanism in the olfactory circuit of a fruit-fly as a locality sensitive hash (Flyhash) and bloom filter (FBF) has been recently proposed and “reprogrammed” for various machine learning tasks such as similarity search, outlier detection and text embeddings. We propose a novel reprogramming of this hash and bloom filter to emulate the canonical nearest neighbor classifier (NNC) in the challenging Federated Learning (FL) setup where training and test data are spread across parties and no data can leave their respective parties. Specifically, we utilize Flyhash and FBF to create the FlyNN classifier, and theoretically establish conditions where FlyNN matches NNC. We show how FlyNN is trained exactly in a FL setup with low communication overhead to produce FlyNNFL, and how it can be differentially private. Empirically, we demonstrate that (i) FlyNN matches NNC accuracy across 70 OpenML datasets, (ii) FlyNNFL training is highly scalable with low communication overhead, providing up to \(8\times\) speedup with \(16\) parties.

Learning binary representations of instances and classes is a classical problem with several high potential applications. In modern settings, the compression of high-dimensional neural representations to low-dimensional binary codes is a challenging task and often require large bit-codes to be accurate. In this work, we propose a novel method for \(\textbf{L}\)earning \(\textbf{L}\)ow-dimensional binary \(\textbf{C}\)odes \((\textbf{LLC})\) for instances as well as classes. Our method does \({\textit{not}}\) require any side-information, like annotated attributes or label meta-data, and learns extremely low-dimensional binary codes (\(\approx 20\) bits for ImageNet-1K). The learnt codes are super-efficient while still ensuring \(\textit{nearly optimal}\) classification accuracy for ResNet50 on ImageNet-1K. We demonstrate that the learnt codes capture intrinsically important features in the data, by discovering an intuitive taxonomy over classes. We further quantitatively measure the quality of our codes by applying it to the efficient image retrieval as well as out-of-distribution (OOD) detection problems. For ImageNet-100 retrieval problem, our learnt binary codes outperform \(16\) bit HashNet using only \(10\) bits and also are as accurate as \(10\) dimensional real representations. Finally, our learnt binary codes can perform OOD detection, out-of-the-box, as accurately as a baseline that needs \(\approx3000\) samples to tune its threshold, while we require \({\textit{none}}\). Code is open-sourced at https://github.com/RAIVNLab/LLC.

Online hashing methods usually learn the hash functions online, aiming to efficiently adapt to the data variations in the streaming environment. However, when the hash functions are updated, the binary codes for the whole database have to be updated to be consistent with the hash functions, resulting in the inefficiency in the online image retrieval process. In this paper, we propose a novel online hashing framework without updating binary codes. In the proposed framework, the hash functions are fixed and a parametric similarity function for the binary codes is learnt online to adapt to the streaming data. Specifically, a parametric similarity function that has a bilinear form is adopted and a metric learning algorithm is proposed to learn the similarity function online based on the characteristics of the hashing methods. The experiments on two multi-label image datasets show that our method is competitive or outperforms the state-of-the-art online hashing methods in terms of both accuracy and efficiency for multi-label image retrieval.

This paper suggests the use of projective clustering based product quantization for improving nearest neighbor and max-inner-product vector search (MIPS) algorithms. We provide anisotropic and quantized variants of projective clustering which outperform previous clustering methods used for this problem such as ScaNN. We show that even with comparable running time complexity, in terms of lookup-multiply-adds, projective clustering produces more quantization centers resulting in more accurate dot-product estimates. We provide thorough experimentation to support our claims.

A wide range of binary analysis applications, such as bug discovery, malware analysis and code clone detection, require recovery of contextual meanings on a binary code. Recently, binary analysis techniques based on machine learning have been proposed to automatically reconstruct the code representation of a binary instead of manually crafting specifics of the analysis algorithm. However, the existing approaches utilizing machine learning are still specialized to solve one domain of problems, rendering recreation of models for different types of binary analysis. In this paper, we propose DeepSemantic utilizing BERT in producing the semantic-aware code representation of a binary code. To this end, we introduce well-balanced instruction normalization that holds rich information for each of instructions yet minimizing an out-of-vocabulary (OOV) problem. DeepSemantic has been carefully designed based on our study with large swaths of binaries. Besides, DeepSemantic leverages the essence of the BERT architecture into re-purposing a pre-trained generic model that is readily available as a one-time processing, followed by quickly applying specific downstream tasks with a fine-tuning process. We demonstrate DeepSemantic with two downstream tasks, namely, binary similarity comparison and compiler provenance (i.e., compiler and optimization level) prediction. Our experimental results show that the binary similarity model outperforms two state-of-the-art binary similarity tools, DeepBinDiff and SAFE, 49.84% and 15.83% on average, respectively.

Self-attention has become increasingly popular in a variety of sequence modeling tasks from natural language processing to recommendation, due to its effectiveness. However, self-attention suffers from quadratic computational and memory complexities, prohibiting its applications on long sequences. Existing approaches that address this issue mainly rely on a sparse attention context, either using a local window, or a permuted bucket obtained by locality-sensitive hashing (LSH) or sorting, while crucial information may be lost. Inspired by the idea of vector quantization that uses cluster centroids to approximate items, we propose LISA (LInear-time Self Attention), which enjoys both the effectiveness of vanilla self-attention and the efficiency of sparse attention. LISA scales linearly with the sequence length, while enabling full contextual attention via computing differentiable histograms of codeword distributions. Meanwhile, unlike some efficient attention methods, our method poses no restriction on casual masking or sequence length. We evaluate our method on four real-world datasets for sequential recommendation. The results show that LISA outperforms the state-of-the-art efficient attention methods in both performance and speed; and it is up to 57x faster and 78x more memory efficient than vanilla self-attention.

Fast k-Nearest Neighbor search over real-valued vector spaces (KNN) is an important algorithmic task for information retrieval and recommendation systems. We present a method for using reduced precision to represent vectors through quantized integer values, enabling both a reduction in the memory overhead of indexing these vectors and faster distance computations at query time. While most traditional quantization techniques focus on minimizing the reconstruction error between a point and its uncompressed counterpart, we focus instead on preserving the behavior of the underlying distance metric. Furthermore, our quantization approach is applied at the implementation level and can be combined with existing KNN algorithms. Our experiments on both open source and proprietary datasets across multiple popular KNN frameworks validate that quantized distance metrics can reduce memory by 60% and improve query throughput by 30%, while incurring only a 2% reduction in recall.

Approximate nearest neighbor search (ANNS) constitutes an important operation in a multitude of applications, including recommendation systems, information retrieval, and pattern recognition. In the past decade, graph-based ANNS algorithms have been the leading paradigm in this domain, with dozens of graph-based ANNS algorithms proposed. Such algorithms aim to provide effective, efficient solutions for retrieving the nearest neighbors for a given query. Nevertheless, these efforts focus on developing and optimizing algorithms with different approaches, so there is a real need for a comprehensive survey about the approaches’ relative performance, strengths, and pitfalls. Thus here we provide a thorough comparative analysis and experimental evaluation of 13 representative graph-based ANNS algorithms via a new taxonomy and fine-grained pipeline. We compared each algorithm in a uniform test environment on eight real-world datasets and 12 synthetic datasets with varying sizes and characteristics. Our study yields novel discoveries, offerings several useful principles to improve algorithms, thus designing an optimized method that outperforms the state-of-the-art algorithms. This effort also helped us pinpoint algorithms’ working portions, along with rule-of-thumb recommendations about promising research directions and suitable algorithms for practitioners in different fields.

Locality Sensitive Hashing (LSH) is an effective method of indexing a set of items to support efficient nearest neighbors queries in high-dimensional spaces. The basic idea of LSH is that similar items should produce hash collisions with higher probability than dissimilar items. We study LSH for (not necessarily convex) polygons, and use it to give efficient data structures for similar shape retrieval. Arkin et al. represent polygons by their “turning function” - a function which follows the angle between the polygon’s tangent and the \( x \)-axis while traversing the perimeter of the polygon. They define the distance between polygons to be variations of the \( L_p \) (for \(p=1,2\)) distance between their turning functions. This metric is invariant under translation, rotation and scaling (and the selection of the initial point on the perimeter) and therefore models well the intuitive notion of shape resemblance. We develop and analyze LSH near neighbor data structures for several variations of the \( L_p \) distance for functions (for \(p=1,2\)). By applying our schemes to the turning functions of a collection of polygons we obtain efficient near neighbor LSH-based structures for polygons. To tune our structures to turning functions of polygons, we prove some new properties of these turning functions that may be of independent interest. As part of our analysis, we address the following problem which is of independent interest. Find the vertical translation of a function \( f \) that is closest in \( L_1 \) distance to a function \( g \). We prove tight bounds on the approximation guarantee obtained by the translation which is equal to the difference between the averages of \( g \) and \( f \).

Kernel Density Estimation (KDE) is a nonparametric method for estimating the shape of a density function, given a set of samples from the distribution. Recently, locality-sensitive hashing, originally proposed as a tool for nearest neighbor search, has been shown to enable fast KDE data structures. However, these approaches do not take advantage of the many other advances that have been made in algorithms for nearest neighbor algorithms. We present an algorithm called Density Estimation from Approximate Nearest Neighbors (DEANN) where we apply Approximate Nearest Neighbor (ANN) algorithms as a black box subroutine to compute an unbiased KDE. The idea is to find points that have a large contribution to the KDE using ANN, compute their contribution exactly, and approximate the remainder with Random Sampling (RS). We present a theoretical argument that supports the idea that an ANN subroutine can speed up the evaluation. Furthermore, we provide a C++ implementation with a Python interface that can make use of an arbitrary ANN implementation as a subroutine for kernel density estimation. We show empirically that our implementation outperforms state of the art implementations in all high dimensional datasets we considered, and matches the performance of RS in cases where the ANN yield no gains in performance.

The high efficiency in computation and storage makes hashing (including binary hashing and quantization) a common strategy in large-scale retrieval systems. To alleviate the reliance on expensive annotations, unsupervised deep hashing becomes an important research problem. This paper provides a novel solution to unsupervised deep quantization, namely Contrastive Quantization with Code Memory (MeCoQ). Different from existing reconstruction-based strategies, we learn unsupervised binary descriptors by contrastive learning, which can better capture discriminative visual semantics. Besides, we uncover that codeword diversity regularization is critical to prevent contrastive learning-based quantization from model degeneration. Moreover, we introduce a novel quantization code memory module that boosts contrastive learning with lower feature drift than conventional feature memories. Extensive experiments on benchmark datasets show that MeCoQ outperforms state-of-the-art methods. Code and configurations are publicly available at https://github.com/gimpong/AAAI22-MeCoQ.

Deep hashing has been widely applied to large-scale image retrieval by encoding high-dimensional data points into binary codes for efficient retrieval. Compared with pairwise/triplet similarity based hash learning, central similarity based hashing can more efficiently capture the global data distribution. For multi-label image retrieval, however, previous methods only use multiple hash centers with equal weights to generate one centroid as the learning target, which ignores the relationship between the weights of hash centers and the proportion of instance regions in the image. To address the above issue, we propose a two-step alternative optimization approach, Instance-weighted Central Similarity (ICS), to automatically learn the center weight corresponding to a hash code. Firstly, we apply the maximum entropy regularizer to prevent one hash center from dominating the loss function, and compute the center weights via projection gradient descent. Secondly, we update neural network parameters by standard back-propagation with fixed center weights. More importantly, the learned center weights can well reflect the proportion of foreground instances in the image. Our method achieves the state-of-the-art performance on the image retrieval benchmarks, and especially improves the mAP by 1.6%-6.4% on the MS COCO dataset.

Many popular machine learning techniques in natural language processing and data mining rely heavily on high-quality text sources. However real-world text datasets contain a significant amount of spelling errors and improperly punctuated variants where the performance of these models would quickly deteriorate. Moreover, real-world, web-scale datasets contain hundreds of millions or even billions of lines of text, where the existing text cleaning tools are prohibitively expensive to execute over and may require an overhead to learn the corrections. In this paper, we present FLAN, a scalable randomized algorithm to clean and canonicalize massive text data. Our algorithm relies on the Jaccard similarity between words to suggest correction results. We efficiently handle the pairwise word-to-word comparisons via Locality Sensitive Hashing (LSH). We also propose a novel stabilization process to address the issue of hash collisions between dissimilar words, which is a consequence of the randomized nature of LSH and is exacerbated by the massive scale of real-world datasets. Compared with existing approaches, our method is more efficient, both asymptotically and in empirical evaluations, and does not rely on additional features, such as lexical/phonetic similarity or word embedding features. In addition, FLAN does not require any annotated data or supervised learning. We further theoretically show the robustness of our algorithm with upper bounds on the false positive and false negative rates of corrections. Our experimental results on real-world datasets demonstrate the efficiency and efficacy of FLAN.

This paper studies the problem of novel category discovery on single- and multi-modal data with labels from different but relevant categories. We present a generic, end-to-end framework to jointly learn a reliable representation and assign clusters to unlabelled data. To avoid over-fitting the learnt embedding to labelled data, we take inspiration from self-supervised representation learning by noise-contrastive estimation and extend it to jointly handle labelled and unlabelled data. In particular, we propose using category discrimination on labelled data and cross-modal discrimination on multi-modal data to augment instance discrimination used in conventional contrastive learning approaches. We further employ Winner-Take-All (WTA) hashing algorithm on the shared representation space to generate pairwise pseudo labels for unlabelled data to better predict cluster assignments. We thoroughly evaluate our framework on large-scale multi-modal video benchmarks Kinetics-400 and VGG-Sound, and image benchmarks CIFAR10, CIFAR100 and ImageNet, obtaining state-of-the-art results.