BRoCoM: A Bayesian Framework for Robust Computing on Memristor Crossbar

Abstract

Memristor crossbar arrays are considered to be a promising platform for neuromorphic computing. To deploy a trained neural network model on memristor crossbars, memristors need to be programmed to the corresponding weight values. In fact, due to device based process variation and noise, deviations of the stored weights from the trained weights are inevitable, thereby causing the degradation of the actual inference performance. This paper proposes a unified Bayesian inference based framework, BRoCoM, which connects device nonidealities and algorithmic training together for robust computing on memristor crossbars. BRoCoM is able to incorporate different levels of nonidealities into prior weight distribution, and transform robustness optimization to Bayesian neural network training, the weights of neural networks are optimized to accommodate uncertainties and minimize inference degradation. Experimental results confirm the capability of the proposed BRoCoM to achieve stable inference performance while tolerating the non-ideal effects of process variation and noise.