Diamond Sampling For Approximate Maximum All-pairs Dot-product (MAD) Search

Ballard Grey, Pinar Ali, Kolda Tamara G., Seshadhri C.. ICDM 2015

[Paper]    

Given two sets of vectors, $A=\{a_1,\dots ,a_m\}$ and $B=\{b_1,\dots ,b_n\}$, our problem is to find the top-$t$ dot products, i.e., the largest $|a_i\cdot b_j|$ among all possible pairs. This is a fundamental mathematical problem that appears in numerous data applications involving similarity search, link prediction, and collaborative filtering. We propose a sampling-based approach that avoids direct computation of all $mn$ dot products. We select diamonds (i.e., four-cycles) from the weighted tripartite representation of $A$ and $B$. The probability of selecting a diamond corresponding to pair $(i,j)$ is proportional to $(a_i\cdot b_j{)}^2$, amplifying the focus on the largest-magnitude entries. Experimental results indicate that diamond sampling is orders of magnitude faster than direct computation and requires far fewer samples than any competing approach. We also apply diamond sampling to the special case of maximum inner product search, and get significantly better results than the state-of-the-art hashing methods.

Similar Work

• Efficient Retrieval With Learned Similarities
• Quantization Based Fast Inner Product Search
• Consistent Subset Sampling
• Weighted Minwise Hashing Beats Linear Sketching For Inner Product Estimation