Direct Learning-Based Deep Spiking Neural Networks: A Review
It provides a comprehensive survey for researchers in neuromorphic computing, but is incremental as it summarizes existing work without new results.
This paper reviews direct learning-based deep spiking neural networks (SNNs), which tackle the optimization difficulty caused by discontinuous spike mechanisms, and highlights progress in accuracy, efficiency, and temporal dynamics utilization methods.
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.