Direct Learning-Based Deep Spiking Neural Networks: A Review

TL;DR

This paper reviews recent advances in direct learning methods for deep spiking neural networks, focusing on accuracy, efficiency, and temporal dynamics, highlighting progress and future challenges in the field.

Contribution

It provides a comprehensive categorization and analysis of direct learning-based deep SNN methods, offering insights into current trends and future research directions.

Findings

Surrogate gradient methods effectively train deep SNNs.

Recent techniques improve accuracy and efficiency of deep SNNs.

Future challenges include optimization and temporal dynamics utilization.

Abstract

The spiking neural network (SNN), as a promising brain-inspired computational model with binary spike information transmission mechanism, rich spatially-temporal dynamics, and event-driven characteristics, has received extensive attention. However, its intricately discontinuous spike mechanism brings difficulty to the optimization of the deep SNN. Since the surrogate gradient method can greatly mitigate the optimization difficulty and shows great potential in directly training deep SNNs, a variety of direct learning-based deep SNN works have been proposed and achieved satisfying progress in recent years. In this paper, we present a comprehensive survey of these direct learning-based deep SNN works, mainly categorized into accuracy improvement methods, efficiency improvement methods, and temporal dynamics utilization methods. In addition, we also divide these categorizations into finer granularities further to better organize and introduce them. Finally, the challenges and trends that may be faced in future research are prospected.