SPIDE: A Purely Spike-based Method for Training Feedback Spiking Neural Networks

Abstract

Spiking neural networks (SNNs) with event-based computation are promising brain-inspired models for energy-efficient applications on neuromorphic hardware. However, most supervised SNN training methods, such as conversion from artificial neural networks or direct training with surrogate gradients, require complex computation rather than spike-based operations of spiking neurons during training. In this paper, we study spike-based implicit differentiation on the equilibrium state (SPIDE) that extends the recently proposed training method, implicit differentiation on the equilibrium state (IDE), for supervised learning with purely spike-based computation, which demonstrates the potential for energy-efficient training of SNNs. Specifically, we introduce ternary spiking neuron couples and prove that implicit differentiation can be solved by spikes based on this design, so the whole training procedure, including both forward and backward passes, is made as event-driven spike computation, and weights are updated locally with two-stage average firing rates. Then we propose to modify the reset membrane potential to reduce the approximation error of spikes. With these key components, we can train SNNs with flexible structures in a small number of time steps and with firing sparsity during training, and the theoretical estimation of energy costs demonstrates the potential for high efficiency. Meanwhile, experiments show that even with these constraints, our trained models can still achieve competitive results on MNIST, CIFAR-10, CIFAR-100, and CIFAR10-DVS. Our code is available at https://github.com/pkuxmq/SPIDE-FSNN.

Accepted by Neural Networks

References in corpus (13)

• High-Performance Large-Scale Image Recognition Without Normalization
• Deep Equilibrium Models
• Direct Feedback Alignment Provides Learning in Deep Neural Networks
• Temporal Efficient Training of Spiking Neural Network via Gradient Re-weighting
• Optimal Conversion of Conventional Artificial Neural Networks to Spiking Neural Networks
• Optimal ANN-SNN Conversion for High-accuracy and Ultra-low-latency Spiking Neural Networks
• Enabling Deep Spiking Neural Networks with Hybrid Conversion and Spike Timing Dependent Backpropagation
• A Free Lunch From ANN: Towards Efficient, Accurate Spiking Neural Networks Calibration
• Characterizing signal propagation to close the performance gap in unnormalized ResNets
• Towards deep learning with spiking neurons in energy based models with contrastive Hebbian plasticity
• A Statistical Framework for Low-bitwidth Training of Deep Neural Networks
• SpikeGrad: An ANN-equivalent Computation Model for Implementing Backpropagation with Spikes
• Stabilizing Equilibrium Models by Jacobian Regularization