Theory and Tools for the Conversion of Analog to Spiking Convolutional Neural Networks

Abstract

Deep convolutional neural networks (CNNs) have shown great potential for numerous real-world machine learning applications, but performing inference in large CNNs in real-time remains a challenge. We have previously demonstrated that traditional CNNs can be converted into deep spiking neural networks (SNNs), which exhibit similar accuracy while reducing both latency and computational load as a consequence of their data-driven, event-based style of computing. Here we provide a novel theory that explains why this conversion is successful, and derive from it several new tools to convert a larger and more powerful class of deep networks into SNNs. We identify the main sources of approximation errors in previous conversion methods, and propose simple mechanisms to fix these issues. Furthermore, we develop spiking implementations of common CNN operations such as max-pooling, softmax, and batch-normalization, which allow almost loss-less conversion of arbitrary CNN architectures into the spiking domain. Empirical evaluation of different network architectures on the MNIST and CIFAR10 benchmarks leads to the best SNN results reported to date.

9 pages, 2 figures, presented at the workshop "Computing with Spikes" at NIPS 2016, Barcelona, Spain

References in corpus (2)

• Striving for Simplicity: The All Convolutional Net
• Fast and Efficient Asynchronous Neural Computation with Adapting Spiking Neural Networks

Cited by in corpus (12)

• Optimal Conversion of Conventional Artificial Neural Networks to Spiking Neural Networks
• Design Space Exploration of Hardware Spiking Neurons for Embedded Artificial Intelligence
• RMP-SNN: Residual Membrane Potential Neuron for Enabling Deeper High-Accuracy and Low-Latency Spiking Neural Network
• DashNet: A Hybrid Artificial and Spiking Neural Network for High-speed Object Tracking
• Efficient Computation in Adaptive Artificial Spiking Neural Networks
• Optimal ANN-SNN Conversion for Fast and Accurate Inference in Deep Spiking Neural Networks
• LIAF-Net: Leaky Integrate and Analog Fire Network for Lightweight and Efficient Spatiotemporal Information Processing
• Algorithm and Hardware Design of Discrete-Time Spiking Neural Networks Based on Back Propagation with Binary Activations
• A Deep Unsupervised Feature Learning Spiking Neural Network with Binarized Classification Layers for EMNIST Classification using SpykeFlow
• STCSNN: High energy efficiency spike-train level spiking neural networks with spatio-temporal conversion
• A Fully Spiking Hybrid Neural Network for Energy-Efficient Object Detection
• RRAM based neuromorphic algorithms