Year: 2022
Type: article
Abstract: This paper presents a novel network compression framework,Kernel Quantization(KQ), targeting to efficiently convert any pre-trained full-precision convolutional neural network (CNN) model into a low-precision version without significant performance loss. Unlike existing methods struggling with weight bit-length,KQhas the potential in improving the compression ratio by considering the convolution kernel as the quantization unit. Inspired by the evolution from weight pruning to filter pruning, we propose to quantize in both kernel and weight level. Instead of representing each weight parameter with a low-bit index, we learn a kernel codebook and replace all kernels in the convolution layer with corresponding low-bit indexes. Thus,KQcan represent the weight tensor in the convolution layer with low-bit indexes and a kernel codebook with limited size, which enablesKQto achieve significant compression ratio. Then, we conduct a 6-bit parameter quantization on the kernel codebook to further reduce redundancy. Extensive experiments on the ImageNet classification task prove thatKQneeds 1.05 and 1.62 bits on average in VGG and ResNet18, respectively, to represent each parameter in the convolution layer and achieves the state-of-the-art compression ratio with little accuracy loss.
Source: IEEE Access
Authors: Zhongzhi Yu, Yemin Shi
Institutions: Peking University, Beijing Academy of Artificial Intelligence
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $1,850
Awards: G5254191062, G8450824501
OpenAlex ID: W4205252534