Thread: Towards fine-grained precision reconfiguration in variable-precision neural network accelerator
Zhang, Shichang1,2; Wang, Ying1; Chen, Xiaoming1; Han, Yinhe1,2; Wang, Yujie1; Li, Xiaowei1,2
刊名IEICE ELECTRONICS EXPRESS
2019-07-25
卷号16期号:14页码:6
关键词DNN accelerator variable bit-precision bit-serial bit-parallel fine-grained precision
ISSN号1349-2543
DOI10.1587/elex.16.20190145
英文摘要In this work, we propose a neural network accelerator that supports finer-grained precision tuning for both activations and weights. To the best of our knowledge, this is the first neural network accelerator supporting arbitrary bit widths within 8 bits for both neural weights and activations. Our accelerator combines the features of bit-parallel and bit-serial design methods so that we can accomplish high arithmetic-unit utilization and precision tuning flexibility simultaneously. According to the cycle accurate simulation and the synthesized result, the proposed accelerator achieves up to 1.77x energy-efficiency improvement over the hit-serial design of Stripes for the evaluated workloads.
资助项目National Natural Science Foundation of China (NSFC)[61834006] ; National Natural Science Foundation of China (NSFC)[61804155] ; National Natural Science Foundation of China (NSFC)[61874124] ; Beijing Municipal Science & Technology Commission[Z171100000117019] ; Beijing Municipal Science & Technology Commission[Z181100008918006] ; Strategic Priority Research Program of the Chinese Academy of Sciences[XDPB12] ; Innovative Project of ICT, CAS[5120186140] ; Youth Innovation Promotion Association CAS ; Young Elite Scientists Sponsorship Program by CAST[2018QNRC001] ; Young Elite Scientists Sponsorship Program by CAST[2016QNRC001]
WOS研究方向Engineering
语种英语
出版者IEICE-INST ELECTRONICS INFORMATION COMMUNICATIONS ENG
WOS记录号WOS:000480417500001
内容类型期刊论文
源URL[http://119.78.100.204/handle/2XEOYT63/4451]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Han, Yinhe
作者单位1.Chinese Acad Sci, Inst Comp Technol, State Key Lab Comp Architecture, Beijing 100190, Peoples R China
2.Univ Chinese Acad Sci, Beijing 100190, Peoples R China
推荐引用方式
GB/T 7714
Zhang, Shichang,Wang, Ying,Chen, Xiaoming,et al. Thread: Towards fine-grained precision reconfiguration in variable-precision neural network accelerator[J]. IEICE ELECTRONICS EXPRESS,2019,16(14):6.
APA Zhang, Shichang,Wang, Ying,Chen, Xiaoming,Han, Yinhe,Wang, Yujie,&Li, Xiaowei.(2019).Thread: Towards fine-grained precision reconfiguration in variable-precision neural network accelerator.IEICE ELECTRONICS EXPRESS,16(14),6.
MLA Zhang, Shichang,et al."Thread: Towards fine-grained precision reconfiguration in variable-precision neural network accelerator".IEICE ELECTRONICS EXPRESS 16.14(2019):6.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。


©版权所有 ©2017 CSpace - Powered by CSpace