DCR: Decomposition-Aware Column Re-Mapping for Stuck-At-Fault Tolerance in ReRAM Arrays

The ReRAM-based neuromorphic computing system (NCS) has been widely used as an energy-efficient platform for deep neural network (DNN) acceleration. However, ReRAM commonly suffers from stuck-at-fault (SAF), resulting in permanent device failure. SAF tolerance is an essential task to ensure the reliability of the system by minimizing the DNN inference accuracy degradation. Since hardware-based solutions incur additional overhead and power consumption, it is necessary to seek a solution that can be executed offline to mitigate the impact of SAF. In this work, we propose a decomposition-aware column re-mapping (DCR) for SAF tolerance in analog ReRAM arrays (RAs). Our DCR consists of the column re-mapping technique combined with fault-aware weight decomposition and an advanced sensitivity metric. As a result, it generates a final weight map optimized for the fault map. Our DCR achieves only about 1% loss of inference accuracy on CIFAR-10 and CIFAR-100 for the analog RAs with the SAF rate of 2% and 1%, respectively, without any hardware-based solution or re-training.

I. Introduction

The neuromorphic computing system (NCS) is an emerging technology that implements the deep neural network (DNN) with high energy efficiency and performance [1]. To fully realize the potential of the NCS, resistive random access memory (ReRAM) [2] has gained attention as an advanced memory technology capable of supporting the energy efficiency and high-performance requirements of the system.