Fault-Tolerant Training Enabled by On-Line Fault Detection for RRAM-Based Neural Computing Systems

摘要: An resistive random-access memory (RRAM)-based computing system (RCS) is an attractive hardware platform for implementing neural algorithms. On-line training RCS enables hardware-based learning a given application and reduces the additional error caused by device parameter variations. However, high occurrence rate of hard faults due to immature fabrication processes limited write endurance restrict applicability on-line RCS. We propose fault-tolerant method that alternates between fault-detection phase phase. In phase, quiescent-voltage comparison utilized. threshold-training remapping scheme proposed. Our results show that, compared without fault tolerance, recognition accuracy Cifar-10 dataset improves from 37% 83% when using low-endurance RRAM cells, 63% 76% cells with but percentage initial faults.