Abstract:Neural networks, one of the key artificial intelligence technologies today, have the computational power and learning ability similar to the brain. However, implementation of neural networks based on the CMOS von Neumann computing systems suffers from the communication bottleneck restricted by the bus bandwidth and memory wall resulting from CMOS downscaling. Consequently, applications based on large‐scale neural networks are energy/area hungry and neuromorphic computing systems are proposed for efficient implementation of neural networks. Neuromorphic computing system consists of the synaptic device, neuronal circuit, and neuromorphic architecture. With the two‐terminal nonvolatile nanoscale memristor as the synaptic device and crossbar as parallel architecture, memristor crossbars are proposed as a promising candidate for neuromorphic computing. Herein, neuromorphic computing systems with memristor crossbars are reviewed. The feasibility and applicability of memristor crossbars based neuromorphic computing for the implementation of artificial neural networks and spiking neural networks are discussed and the prospects and challenges are also described. In this paper, the basic principles of synapses and synaptic memristors, neurons, and neuronal memristor, and neuromorphic memristor crossbar architecture are introduced. The feasibility and applicability of memristors crossbar approached neuromorphic computing system for the implementation of artificial neural networks and spiking neural networks are discussed. The prospects and challenges are also described.

Block-Clustering on Neural Networks for Large-scale Memristor-based Implementation

Reliable Memristor-based Neuromorphic Design Using Variation- and Defect-Aware Training

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

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

Memristor Neural Network Design

Memristor-based Neuromorphic Implementations for Artificial Neural Networks

On-Chip Universal Supervised Learning Methods for Neuro-Inspired Block of Memristive Nanodevices

Build Reliable And Efficient Neuromorphic Design With Memristor Technology

Neuromorphic Computing with Memristor Crossbar

Memristors: Associative Learning with Temporal Contiguity in a Memristive Circuit for Large‐Scale Neuromorphic Networks (Adv. Electron. Mater. 8/2015)

In-Memory Computing With Memristor Arrays

The Ouroboros of Memristors: Neural Networks Facilitating Memristor Programming

Exploiting Memristors for Neuromorphic Reinforcement Learning

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.

Efficient and self-adaptive in-situ learning in multilayer memristor neural networks

Associative Learning with Temporal Contiguity in a Memristive Circuit for Large‐Scale Neuromorphic Networks

Challenges of Memristor Based Neuromorphic Computing System

On-chip Supervised Learning Rule for Ultra High Density Neural Crossbar Using Memristor for Synapse and Neuron

Robust Learning Approach for Neuro-Inspired Nanoscale Crossbar Architecture.

Memristor-Based Neuromorphic System With Content Addressable Memory Structure

Overview of Memristor-Based Neural Network Design and Applications