Abstract: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 design framework for a memristive crossbar-based NCS in combination with general and chip-specific design optimizations. First, we design a general reliability-aware training scheme to enhance the robustness of NCS to SAFs and device variations; a dropconnect-inspired approach is 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 proposed to obtain a smooth error curve, where the effects of variations are suppressed. Second, given the neural network model generated by the reliability-aware training scheme, we exploit chip-specific mapping and re- training to further improve computation accuracy loss incurred by SAFs. Experimental results show that the proposed method can boost the computation accuracy of NCS and improve the NCS robustness. Note to Practitioners—This work is motivated by the manufacturing reliability problem in a memristive crossbar-based NCS. To enhance the robustness of an NCS to SAFs and device variations, this paper presents a reliability-driven design framework with taking account of both general and chip-specific design optimizations. The experimental results have demonstrated that the proposed framework is superior to the prior arts, and can be easily integrated with existing industrial hardware-based fault tolerance solutions for higher accuracy at lower overhead. Memristive crossbar-based computing system gives hope for the anticipated efficient implementation of artificial neuromorphic networks. With the help of the reliability-driven designs, the computation accuracy is restored, and hence we can expect the wide use of memristive crossbar-based computing system in neuromorphic computing applications.

Memristive Crossbar Mapping for Neuromorphic Computing Systems on 3D IC.

BRoCoM: A Bayesian Framework for Robust Computing on Memristor Crossbar.

Aging Aware Retraining for Memristor-based Neuromorphic Computing

Neuromorphic Computing with Memristor Crossbar

Modeling and Analysis of a Large-Scale Memristor Crossbar Array for Neuromorphic Chip

An Efficient Programming Framework for Memristor-based Neuromorphic Computing

Latency Insertion Method for Accelerated Simulation of Memristor Crossbar Array in Neuromorphic Chip

Novel Three-Dimensional Artificial Neural Network Based on an Eight-Layer Vertical Memristor with an Ultrahigh Rectify Ratio (>107) and an Ultrahigh Nonlinearity (>105) for Neuromorphic Computing.

Memristive Crossbar Arrays for Storage and Computing Applications

A Brain-Inspired In-Memory Computing System for Neuronal Communication Via Memristive Circuits.

An Eda Framework For Large Scale Hybrid Neuromorphic Computing Systems

An Overview On Memristor Crossabr Based Neuromorphic Circuit And Architecture

Approaching the mapping limit with closed-loop mapping strategy for deploying neural networks on neuromorphic hardware

Reliability-Driven Memristive Crossbar Design in Neuromorphic Computing Systems

Fault-Tolerant-Driven Clustering for Large Scale Neuromorphic Computing Systems

Modeling and Design of A 3D Interconnect Based Circuit Cell Formed with 3D SiP Techniques Mimicking Brain Neurons for Neuromorphic Computing Applications

A Spiking Neuromorphic Design with Resistive Crossbar

Scaling Limits of Memristor-Based Routers for Asynchronous Neuromorphic Systems

Memristor Neural Network Design

Memcapacitor Crossbar Array with Charge Trap NAND Flash Structure for Neuromorphic Computing

A Multi-Value 3D Crossbar Array Nonvolatile Memory Based on Pure Memristors