Flexor Trainable Fractional Quantization

Dongsoo Lee, Se Jung Kwon, Byeongwook Kim, Yongkweon Jeon, Baeseong Park, Jeongin Yun. Neural Information Processing Systems 2020

[Paper]    
NEURIPS Quantisation Supervised

Quantization based on the binary codes is gaining attention because each quantized bit can be directly utilized for computations without dequantization using look-up tables. Previous attempts, however, only allow for integer numbers of quantization bits, which ends up restricting the search space for compression ratio and accuracy. In this paper, we propose an encryption algorithm/architecture to compress quantized weights so as to achieve fractional numbers of bits per weight. Decryption during inference is implemented by digital XOR-gate networks added into the neural network model while XOR gates are described by utilizing $\tanh (x)$ for backward propagation to enable gradient calculations. We perform experiments using MNIST, CIFAR-10, and ImageNet to show that inserting XOR gates learns quantization/encrypted bit decisions through training and obtains high accuracy even for fractional sub 1-bit weights. As a result, our proposed method yields smaller size and higher model accuracy compared to binary neural networks.

Similar Work

• Focused Quantization For Sparse Cnns
• From Hashing To Cnns Training Binaryweight Networks Via Hashing
• Compositional Coding For Collaborative Filtering
• Accurate Deep Representation Quantization With Gradient Snapping Layer For Similarity Search