paper-with-me

Papers

Neural Network Activation Quantization with Bitwise Information Bottlenecks

2020-06-09 · Xichuan Zhou, Kui Liu, Cong Shi, Haijun Liu, Ji Liu

Recent researches on information bottleneck shed new light on the continuous attempts to open the black box of neural signal encoding. Inspired by the problem of lossy signal compression for wireless communication, this paper presents a Bitwise Information Bottleneck approach for quantizing and encoding neural network activations. Based on the rate-distortion theory, the Bitwise Information Bottleneck attempts to determine the most significant bits in activation representation by assigning and approximating the sparse coefficient associated with each bit. Given the constraint of a limited average code rate, the information bottleneck minimizes the rate-distortion for optimal activation quantization in a flexible layer-by-layer manner. Experiments over ImageNet and other datasets show that, by minimizing the quantization rate-distortion of each layer, the neural network with information bottlenecks achieves the state-of-the-art accuracy with low-precision activation. Meanwhile, by reducing the code rate, the proposed method can improve the memory and computational efficiency by over six times compared with the deep neural network with standard single-precision representation. Codes will be available on GitHub when the paper is accepted \url{https://github.com/BitBottleneck/PublicCode}.

📄 PDF Abstract BibTeX arXiv:2006.05210

Code (1)

CQUlearningsystemgroup/BitwiseBottleneck 공식 구현 tf

Tasks

Computational EfficiencyQuantization

Similar Papers 제목 키워드 기반

Effective Quantization Methods for Recurrent Neural Networks

2016-11-30 · Qinyao He, He Wen, Shuchang Zhou, Yuxin Wu 외

Reducing bit-widths of weights, activations, and gradients of a Neural Network can shrink its storage size and memory usage, and also allow for faster training and inference by exploiting bitwise operations. However, pre…

Quantization

OPTIMAL BINARY QUANTIZATION FOR DEEP NEURAL NETWORKS

2019-09-25 · Hadi Pouransari, Oncel Tuzel

Quantizing weights and activations of deep neural networks results in significant improvement in inference efficiency at the cost of lower accuracy. A source of the accuracy gap between full precision and quantized model…

Quantization

Least squares binary quantization of neural networks

2020-01-09 · Hadi Pouransari, Zhucheng Tu, Oncel Tuzel

Quantizing weights and activations of deep neural networks results in significant improvement in inference efficiency at the cost of lower accuracy. A source of the accuracy gap between full precision and quantized model…

Quantization

Forward and Backward Information Retention for Accurate Binary Neural Networks

2019-09-24 · CVPR 2020 6 · Haotong Qin, Ruihao Gong, Xianglong Liu, Mingzhu Shen 외

Weight and activation binarization is an effective approach to deep neural network compression and can accelerate the inference by leveraging bitwise operations. Although many binarization methods have improved the accur…

BinarizationNeural Network CompressionQuantization

Semi-Relaxed Quantization with DropBits: Training Low-Bit Neural Networks via Bitwise Regularization

2021-01-01 · Jung Hyun Lee, Jihun Yun, Sung Ju Hwang, Eunho Yang

Network quantization, which aims to reduce the bit-lengths of the network weights and activations, has emerged as one of the key ingredients to reduce the size of neural networks for their deployments to resource-limited…

Quantization