Abstract:Spiking Neural Networks (SNNs) are a pathway that could potentially empower low-power event-driven neuromorphic hardware due to their spatio-temporal information processing capability and high biological plausibility. Although SNNs are currently more efficient than artificial neural networks (ANNs), they are not as accurate as ANNs. Error backpropagation is the most common method for directly training neural networks, promoting the prosperity of ANNs in various deep learning fields. However, since the signals transmitted in the SNN are non-differentiable discrete binary spike events, the activation function in the form of spikes presents difficulties for the gradient-based optimization algorithms to be directly applied in SNNs, leading to a performance gap (i.e., accuracy and latency) between SNNs and ANNs. This paper introduces a new learning algorithm, called SSTDP, which bridges the gap between backpropagation (BP)-based learning and spike-time-dependent plasticity (STDP)-based learning to train SNNs efficiently. The scheme incorporates the global optimization process from BP and the efficient weight update derived from STDP. It not only avoids the non-differentiable derivation in the BP process but also utilizes the local feature extraction property of STDP. Consequently, our method can lower the possibility of vanishing spikes in BP training and reduce the number of time steps to reduce network latency. In SSTDP, we employ temporal-based coding and use Integrate-and-Fire (IF) neuron as the neuron model to provide considerable computational benefits. Our experiments show the effectiveness of the proposed SSTDP learning algorithm on the SNN by achieving the best classification accuracy 99.3% on the Caltech 101 dataset, 98.1% on the MNIST dataset, and 91.3% on the CIFAR-10 dataset compared to other SNNs trained with other learning methods. It also surpasses the best inference accuracy of the directly trained SNN with 25~32× less inference latency. Moreover, we analyze event-based computations to demonstrate the efficacy of the SNN for inference operation in the spiking domain, and SSTDP methods can achieve 1.3~37.7× fewer addition operations per inference. The code is available at: https://github.com/MXHX7199/SNN-SSTDP .

Graph-Based Spatio-Temporal Backpropagation for Training Spiking Neural Networks

Spatio-Temporal Backpropagation for Training High-Performance Spiking Neural Networks

Spatiotemporal Backpropagation based on Channel Reward for Training High-Precision Spiking Neural Network

SSTDP: Supervised Spike Timing Dependent Plasticity for Efficient Spiking Neural Network Training

Backpropagation with Sparsity Regularization for Spiking Neural Network Learning.

An Efficient Non-Backpropagation Method for Training Spiking Neural Networks

Spike-Train Level Backpropagation for Training Deep Recurrent Spiking Neural Networks

Direct Training via Backpropagation for Ultra-low Latency Spiking Neural Networks with Multi-threshold

An Efficient Learning Algorithm for Direct Training Deep Spiking Neural Networks

Spike Trains Encoding and Threshold Rescaling Method for Deep Spiking Neural Networks

A temporally and spatially local spike-based backpropagation algorithm to enable training in hardware

BP-STDP: Approximating backpropagation using spike timing dependent plasticity

Deep Spiking Neural Network Using Spatio-temporal Backpropagation with Variable Resistance.

Analyzing and Accelerating the Bottlenecks of Training Deep SNNs With Backpropagation

Efficient Spiking Neural Networks with Logarithmic Temporal Coding.

Learnable Surrogate Gradient for Direct Training Spiking Neural Networks

Adaptive Multi-Level Firing for Direct Training Deep Spiking Neural Networks

Training Spiking Neural Networks with Event-driven Backpropagation

Enabling Spike-Based Backpropagation for Training Deep Neural Network Architectures

Towards Memory- and Time-Efficient Backpropagation for Training Spiking Neural Networks

Spiking Deep Residual Networks.