Reliability-Driven Neural Network Training for Memristive Crossbar-Based Neuromorphic Computing Systems

Wang, Junpeng1,2; Xu, Qi3; Yuan, Bo4; Chen, Song1,2; Yu, Bei5; Wu, Feng1,2

10.1109/ISCAS45731.2020.9180923

2020

IEEE International Symposium on Circuits and Systems (ISCAS)

0271-4302

978-1-7281-3320-1

2020 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS (ISCAS)

2020-October

1-4

OCT 10-21, 2020

null,null,ELECTR NETWORK

345 E 47TH ST, NEW YORK, NY 10017 USA

IEEE

In recent years, memristive crossbar-based neuromorphic computing systems (NCS) have provided a promising solution to the acceleration of neural networks. However, stuck-at faults (SAFs) in the memristor devices significantly degrade the computing accuracy of NCS. Besides, the memristor suffers from the process variations, causing deviation of the actual programming resistance from its target resistance. In this paper, we propose a reliability-driven network training framework for a memristive crossbar-based NCS, with taking account of both SAFs and device variations challenges. A dropout-inspired approach is first developed to alleviate the impact of SAFs. A new weighted error function, including cross-entropy error (CEE), the l(2)-norm of weights, and the sum of squares of first-order derivatives of CEE with respect to weights, is further proposed to obtain a smooth error curve, where the effects of variations are suppressed. Experimental results show that the proposed method can boost the computation accuracy of NCS and improve the NCS robustness.

neuromorphic computing system memristive crossbar fault tolerance robustness

其他

英语

CPCI-S ; EI

National Natural Science Foundation of China (NSFC)[61874102,61732020,61904047]

Engineering

Engineering, Electrical & Electronic

WOS:000706854700106

20212810618761

Errors ; Memristors

Semiconductor Devices and Integrated Circuits:714.2

Web of Science

1.Univ Sci & Technol China, Sch Microelect, Beijing, Peoples R China
2.USTC Beijing Res Inst, Beijing, Peoples R China
3.Hefei Univ Technol, Dept Elect Sci & Technol, Hefei, Anhui, Peoples R China
4.Southern Univ Sci & Technol, Dept Comp Sci & Engn, Shenzhen, Guangdong, Peoples R China
5.Chinese Univ Hong Kong, Dept Comp Sci & Engn, Hong Kong, Peoples R China

Wang, Junpeng,Xu, Qi,Yuan, Bo,et al. Reliability-Driven Neural Network Training for Memristive Crossbar-Based Neuromorphic Computing Systems[C]. 345 E 47TH ST, NEW YORK, NY 10017 USA:IEEE,2020:1-4.