Abstract:Artificial neural networks (ANNs) have led to landmark changes in many fields, but they still differ significantly from the mechanisms of real biological neural networks and face problems such as high computing costs, excessive computing power, and so on. Spiking neural networks (SNNs) provide a new approach combined with brain-like science to improve the computational energy efficiency, computational architecture, and biological credibility of current deep learning applications. In the early stage of development, its poor performance hindered the application of SNNs in real-world scenarios. In recent years, SNNs have made great progress in computational performance and practicability compared with the earlier research results, and are continuously producing significant results. Although there are already many pieces of literature on SNNs, there is still a lack of comprehensive review on SNNs from the perspective of improving performance and practicality as well as incorporating the latest research results. Starting from this issue, this paper elaborates on SNNs along the complete usage process of SNNs including network construction, data processing, model training, development, and deployment, aiming to provide more comprehensive and practical guidance to promote the development of SNNs. Therefore, the connotation and development status of SNN computing is reviewed systematically and comprehensively from four aspects: composition structure, data set, learning algorithm, software/hardware development platform. Then the development characteristics of SNNs in intelligent computing are summarized, the current challenges of SNNs are discussed and the future development directions are also prospected. Our research shows that in the fields of machine learning and intelligent computing, SNNs have comparable network scale and performance to ANNs and the ability to challenge large datasets and a variety of tasks. The advantages of SNNs over ANNs in terms of energy efficiency and spatial-temporal data processing have been more fully exploited. And the development of programming and deployment tools has lowered the threshold for the use of SNNs. SNNs show a broad development prospect for brain-like computing.

Recent Advances and New Frontiers in Spiking Neural Networks

A Survey of Neuromorphic Computing Based on Spiking Neural Networks

A Review of Recent Advances and Application for Spiking Neural Networks

Research Advances and New Paradigms for Biology-inspired Spiking Neural Networks

A Review of Spiking Neural Networks

Editorial: Spiking Neural Network Learning, Benchmarking, Programming and Executing

A Short Survey of the Development and Applications of Spiking Neural Networks of High Biological Plausibility

Advanced Computing and Related Applications Leveraging Brain-inspired Spiking Neural Networks

A Review of Computing with Spiking Neural Networks

Advancements in Algorithms and Neuromorphic Hardware for Spiking Neural Networks

Spiking Neural Networks for Computational Intelligence: An Overview

Direct Learning-Based Deep Spiking Neural Networks: A Review

An Introductory Review of Spiking Neural Network and Artificial Neural Network: From Biological Intelligence to Artificial Intelligence

SNN Learning, Benchmarking, Programming, and Executing

CSNN: an Augmented Spiking Based Framework with Perceptron-Inception

A review of learning in biologically plausible spiking neural networks

Deep CovDenseSNN: A Hierarchical Event-Driven Dynamic Framework with Spiking Neurons in Noisy Environment

[A Review of Brain-Like Spiking Neural Network and Its Neuromorphic Chip Research].

Recent Advances in Scalable Energy-Efficient and Trustworthy Spiking Neural networks: from Algorithms to Technology

Research Progress of spiking neural network in image classification: a review

Brain-inspired Evolutionary Architectures for Spiking Neural Networks