Extremely Sparse Networks via Binary Augmented Pruning for Fast Image Classification

doi:10.1109/TNNLS.2021.3120409

	Extremely Sparse Networks via Binary Augmented Pruning for Fast Image Classification
	Wang, Peisong1 ; Li, Fanrong2 ; Li, Gang3 ; Cheng, Jian1
发表期刊	IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
ISSN	2162-237X
	2021-11-08
页码	14
通讯作者	Cheng, Jian(jcheng@nlpr.ia.ac.cn)
摘要	Network pruning and binarization have been demonstrated to be effective in neural network accelerator design for high speed and energy efficiency. However, most existing pruning approaches achieve a poor tradeoff between accuracy and efficiency, which on the other hand, has limited the progress of neural network accelerators. At the same time, binary networks are highly efficient, however, a large accuracy gap exists between binary networks and their full-precision counterparts. In this article, we investigate the merits of extremely sparse networks with binary connections for image classification through software-hardware codesign. More specifically, we first propose a binary augmented extremely pruning method that can achieve similar to 98% sparsity with small accuracy degradation. Then we design the hardware architecture based on the resulting sparse and binary networks, which extensively explores the benefits of extreme sparsity with negligible resource consumption introduced by binary branch. Experiments on large-scale ImageNet classification and field-programmable gate array (FPGA) demonstrate that the proposed software-hardware architecture can achieve a prominent tradeoff between accuracy and efficiency.
关键词	Hardware acceleration image classification neural networks pruning software-hardware codesign
DOI	10.1109/TNNLS.2021.3120409
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[61906193] ; Strategic Priority Research Program of Chinese Academy of Sciences[XDA27040300] ; National Key Research and Development Program of China[2020AAA0103402]
项目资助者	National Natural Science Foundation of China ; Strategic Priority Research Program of Chinese Academy of Sciences ; National Key Research and Development Program of China
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Computer Science, Hardware & Architecture ; Computer Science, Theory & Methods ; Engineering, Electrical & Electronic
WOS记录号	WOS:000732358700001
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
七大方向——子方向分类	AI芯片与智能计算
引用统计	被引频次：2[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/46978
专题	复杂系统认知与决策实验室_高效智能计算与学习
通讯作者	Cheng, Jian
作者单位	1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 2.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 3.Shanghai Jiao Tong Univ, Adv Comp Architecture Lab, Dept Comp Sci & Engn, Shanghai 200240, Peoples R China
第一作者单位	模式识别国家重点实验室
通讯作者单位	模式识别国家重点实验室
推荐引用方式 GB/T 7714	Wang, Peisong,Li, Fanrong,Li, Gang,et al. Extremely Sparse Networks via Binary Augmented Pruning for Fast Image Classification[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,2021:14.
APA	Wang, Peisong,Li, Fanrong,Li, Gang,&Cheng, Jian.(2021).Extremely Sparse Networks via Binary Augmented Pruning for Fast Image Classification.IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,14.
MLA	Wang, Peisong,et al."Extremely Sparse Networks via Binary Augmented Pruning for Fast Image Classification".IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS (2021):14.