Direct Training for Spiking Neural Networks: Faster, Larger, Better

Abstract

Spiking neural networks (SNNs) that enables energy efficient implementation on emerging neuromorphic hardware are gaining more attention. Yet now, SNNs have not shown competitive performance compared with artificial neural networks (ANNs), due to the lack of effective learning algorithms and efficient programming frameworks. We address this issue from two aspects: (1) We propose a neuron normalization technique to adjust the neural selectivity and develop a direct learning algorithm for deep SNNs. (2) Via narrowing the rate coding window and converting the leaky integrate-and-fire (LIF) model into an explicitly iterative version, we present a Pytorch-based implementation method towards the training of large-scale SNNs. In this way, we are able to train deep SNNs with tens of times speedup. As a result, we achieve significantly better accuracy than the reported works on neuromorphic datasets (N-MNIST and DVS-CIFAR10), and comparable accuracy as existing ANNs and pre-trained SNNs on non-spiking datasets (CIFAR10). {To our best knowledge, this is the first work that demonstrates direct training of deep SNNs with high performance on CIFAR10, and the efficient implementation provides a new way to explore the potential of SNNs.

arXiv comment10 pages, 8 figures. Accepted by AAAI 2019

Cited by in corpus (38)

• Enabling Spike-based Backpropagation for Training Deep Neural Network Architectures
• Temporal Spike Sequence Learning via Backpropagation for Deep Spiking Neural Networks
• Optimal Conversion of Conventional Artificial Neural Networks to Spiking Neural Networks
• DIET-SNN: Direct Input Encoding With Leakage and Threshold Optimization in Deep Spiking Neural Networks
• Spiking Deep Residual Network
• STSC-SNN: Spatio-Temporal Synaptic Connection with Temporal Convolution and Attention for Spiking Neural Networks
• Spiking-YOLO: Spiking Neural Network for Energy-Efficient Object Detection
• Enabling Deep Spiking Neural Networks with Hybrid Conversion and Spike Timing Dependent Backpropagation
• ES-ImageNet: A Million Event-Stream Classification Dataset for Spiking Neural Networks
• Training Deep Spiking Neural Networks
• Spiking neural networks trained via proxy
• A Tandem Learning Rule for Effective Training and Rapid Inference of Deep Spiking Neural Networks
• Temporal-Coded Deep Spiking Neural Network with Easy Training and Robust Performance
• Exploring Adversarial Attack in Spiking Neural Networks with Spike-Compatible Gradient
• Comparing SNNs and RNNs on Neuromorphic Vision Datasets: Similarities and Differences
• Progressive Tandem Learning for Pattern Recognition with Deep Spiking Neural Networks
• Optimal ANN-SNN Conversion for Fast and Accurate Inference in Deep Spiking Neural Networks
• Comprehensive SNN Compression Using ADMM Optimization and Activity Regularization
• Towards Low-Latency Energy-Efficient Deep SNNs via Attention-Guided Compression
• VSA: Reconfigurable Vectorwise Spiking Neural Network Accelerator
• LIAF-Net: Leaky Integrate and Analog Fire Network for Lightweight and Efficient Spatiotemporal Information Processing
• Exploiting Neuron and Synapse Filter Dynamics in Spatial Temporal Learning of Deep Spiking Neural Network
• A Spike Learning System for Event-driven Object Recognition
• Training Feedback Spiking Neural Networks by Implicit Differentiation on the Equilibrium State
• Strategy and Benchmark for Converting Deep Q-Networks to Event-Driven Spiking Neural Networks
• Training Energy-Efficient Deep Spiking Neural Networks with Time-to-First-Spike Coding
• HYPER-SNN: Towards Energy-efficient Quantized Deep Spiking Neural Networks for Hyperspectral Image Classification
• You Only Spike Once: Improving Energy-Efficient Neuromorphic Inference to ANN-Level Accuracy
• EventDrop: data augmentation for event-based learning
• Multi-Tones' Phase Coding (MTPC) of Interaural Time Difference by Spiking Neural Network
• Event-based Synthetic Aperture Imaging with a Hybrid Network
• TactileSGNet: A Spiking Graph Neural Network for Event-based Tactile Object Recognition
• One Timestep is All You Need: Training Spiking Neural Networks with Ultra Low Latency
• Ensembles of Spiking Neural Networks
• Minibatch Processing in Spiking Neural Networks
• Spiking Neural Networks with Single-Spike Temporal-Coded Neurons for Network Intrusion Detection
• SiamSNN: Siamese Spiking Neural Networks for Energy-Efficient Object Tracking
• Spiking CapsNet: A Spiking Neural Network With A Biologically Plausible Routing Rule Between Capsules