Temporal-Coded Deep Spiking Neural Network with Easy Training and Robust Performance
This work addresses the challenge of making SNNs practical for energy-efficient AI applications, though it appears incremental by building on existing temporal-coding methods.
The paper tackled the difficult training and lack of hardware platforms for spiking neural networks (SNNs) by developing a temporal-coded deep SNN that achieves testing accuracy within 1% of equivalent deep neural networks on CIFAR10 and ImageNet, with robustness demonstrated through weight quantization to 8, 4, 2 bits and noise perturbation, and a circuit schematic offering 90% energy efficiency gain.
Spiking neural network (SNN) is interesting both theoretically and practically because of its strong bio-inspiration nature and potentially outstanding energy efficiency. Unfortunately, its development has fallen far behind the conventional deep neural network (DNN), mainly because of difficult training and lack of widely accepted hardware experiment platforms. In this paper, we show that a deep temporal-coded SNN can be trained easily and directly over the benchmark datasets CIFAR10 and ImageNet, with testing accuracy within 1% of the DNN of equivalent size and architecture. Training becomes similar to DNN thanks to the closed-form solution to the spiking waveform dynamics. Considering that SNNs should be implemented in practical neuromorphic hardwares, we train the deep SNN with weights quantized to 8, 4, 2 bits and with weights perturbed by random noise to demonstrate its robustness in practical applications. In addition, we develop a phase-domain signal processing circuit schematic to implement our spiking neuron with 90% gain of energy efficiency over existing work. This paper demonstrates that the temporal-coded deep SNN is feasible for applications with high performance and high energy efficient.