Reliability-driven neural network training for memristive crossbar-based neuromorphic computing systems

Wang，Junpeng1; Xu，Qi2; Yuan，Bo3; Chen，Song1; Yu，Bei4; Wu，Feng1

2020

0271-4310

Proceedings - IEEE International Symposium on Circuits and Systems

2020-October

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-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.

Fault tolerance Memristive crossbar Neuromorphic computing system Robustness

其他

英语

EI

20212810618761

Errors ; Memristors

Semiconductor Devices and Integrated Circuits:714.2

2-s2.0-85109268438

Scopus

1.School of Microelectronics,University of Science and Technology of China,USTC Beijing Research Institute,Beijing,China
2.Department of Electronic Science and Technology,Hefei University of Technology,China
3.Department of Computer Science and Engineering,Southern University of Science and Technology,China
4.Department of Computer Science and Engineering,The Chinese University of Hong Kong,Hong Kong

Wang，Junpeng,Xu，Qi,Yuan，Bo,et al. Reliability-driven neural network training for memristive crossbar-based neuromorphic computing systems[C],2020.