中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 计算技术研究所
  3. 中国科学院计算技术研究所
  4. 中国科学院计算技术研究所期刊论文
  5. 英文
Exploring Winograd Convolution for Cost-Effective Neural Network Fault Tolerance

文献类型：期刊论文

作者Xue, Xinghua3,4; Liu, Cheng3,4; Liu, Bo2; Huang, Haitong3,4; Wang, Ying3,4; Luo, Tao1; Zhang, Lei3,4; Li, Huawei3,4; Li, Xiaowei3,4
刊名IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS
出版日期2023-11-01
卷号31期号:11页码:1763-1773
关键词Fault tolerant systems Fault tolerance Artificial neural networks Convolution Reliability Computational modeling Neurons Fault-tolerance soft errors vulnerability analysis winograd convolution (WG-Conv)
ISSN号1063-8210
DOI10.1109/TVLSI.2023.3306894
英文摘要Winograd is generally utilized to optimize convolution performance and computational efficiency because of the reduced multiplication operations, but the reliability issues brought by winograd are usually overlooked. In this work, we observe the great potential of winograd convolution (WG-Conv) in improving neural network (NN) fault tolerance. Based on the observation, we evaluate WG-Conv fault tolerance comprehensively from different granularities ranging from models, layers, and operation types for the first time. Then, we explore the use of inherent fault tolerance of WG-Conv for cost-effective NN protection against soft errors. Specifically, we mainly investigate how WG-Conv can be effectively incorporated with classical fault-tolerant design approaches including triple modular redundancy (TMR), fault-aware retraining, and constrained activation functions. According to our experiments, WG-Conv can reduce the fault-tolerant design overhead by 55.77% on average without any accuracy loss compared to standard convolution (ST-Conv), and further reduce the computing overhead by 17.24% when the inherent fault tolerance of WG-Conv is considered. When it is applied on fault-tolerant NNs enhanced with fault-aware retraining and constrained activation functions, the resulting model accuracy generally shows significant improvement in the presence of various faults.
资助项目National Natural Science Foundation of China[62174162] ; Space Trusted Computing and Electronic Information Technology Laboratory of Beijing Institute of Control Engineering (BICE)[OBCandETL- 2022-07]
WOS研究方向Computer Science ; Engineering
语种英语
WOS记录号WOS:001179765700002
出版者IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
源URL[http://119.78.100.204/handle/2XEOYT63/38820]  
专题中国科学院计算技术研究所期刊论文_英文
通讯作者Liu, Cheng
作者单位1.ASTAR, Inst High Performance Comp, Singapore 138632, Singapore
2.Beijing Inst Control Engn, Beijing 100190, Peoples R China
3.Univ Chinese Acad Sci, Sch Comp Sci & Technol, Beijing 100190, Peoples R China
4.Chinese Acad Sci, Inst Comp Technol, State Key Lab Processors, Beijing 100190, Peoples R China
推荐引用方式
GB/T 7714		 Xue, Xinghua,Liu, Cheng,Liu, Bo,et al. Exploring Winograd Convolution for Cost-Effective Neural Network Fault Tolerance[J]. IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS,2023,31(11):1763-1773.
APA Xue, Xinghua.,Liu, Cheng.,Liu, Bo.,Huang, Haitong.,Wang, Ying.,...&Li, Xiaowei.(2023).Exploring Winograd Convolution for Cost-Effective Neural Network Fault Tolerance.IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS,31(11),1763-1773.
MLA Xue, Xinghua,et al."Exploring Winograd Convolution for Cost-Effective Neural Network Fault Tolerance".IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS 31.11(2023):1763-1773.

入库方式： OAI收割

来源：计算技术研究所

浏览0
下载0
收藏0
其他版本

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。