Spiking-GAN: A Spiking Generative Adversarial Network Using Time-To-First-Spike Coding

TL;DR

This paper introduces Spiking-GAN, a novel spike-based generative adversarial network utilizing time-to-first-spike coding, demonstrating efficient training and high-quality sample generation on MNIST.

Contribution

It is the first spike-based GAN employing temporal coding and approximate backpropagation, enhancing efficiency and sample quality in spiking neural networks.

Findings

Over 33% faster inference with 'Aggressive TTFS' loss

Reduced spikes by more than 11% compared to prior methods

Successfully generated high-quality MNIST samples

Abstract

Spiking Neural Networks (SNNs) have shown great potential in solving deep learning problems in an energy-efficient manner. However, they are still limited to simple classification tasks. In this paper, we propose Spiking-GAN, the first spike-based Generative Adversarial Network (GAN). It employs a kind of temporal coding scheme called time-to-first-spike coding. We train it using approximate backpropagation in the temporal domain. We use simple integrate-and-fire (IF) neurons with very high refractory period for our network which ensures a maximum of one spike per neuron. This makes the model much sparser than a spike rate-based system. Our modified temporal loss function called 'Aggressive TTFS' improves the inference time of the network by over 33% and reduces the number of spikes in the network by more than 11% compared to previous works. Our experiments show that on training the network on the MNIST dataset using this approach, we can generate high quality samples. Thereby demonstrating the potential of this framework for solving such problems in the spiking domain.