Supervised learning based on temporal coding in spiking neural networks

TL;DR

This paper demonstrates that feedforward spiking neural networks using temporal coding have differentiable input-output relations, enabling the transfer of gradient-based training methods from artificial neural networks to spiking networks, which can process complex temporal spike patterns.

Contribution

It introduces a novel approach for training temporally coded spiking neural networks using gradient descent, bridging the gap with traditional ANNs and enabling complex temporal information processing.

Findings

Networks encode information in spike times, not rates.

The input-output relation is differentiable almost everywhere.

Networks spike more sparsely than rate-based models.

Abstract

Gradient descent training techniques are remarkably successful in training analog-valued artificial neural networks (ANNs). Such training techniques, however, do not transfer easily to spiking networks due to the spike generation hard non-linearity and the discrete nature of spike communication. We show that in a feedforward spiking network that uses a temporal coding scheme where information is encoded in spike times instead of spike rates, the network input-output relation is differentiable almost everywhere. Moreover, this relation is piece-wise linear after a transformation of variables. Methods for training ANNs thus carry directly to the training of such spiking networks as we show when training on the permutation invariant MNIST task. In contrast to rate-based spiking networks that are often used to approximate the behavior of ANNs, the networks we present spike much more sparsely and their behavior can not be directly approximated by conventional ANNs. Our results highlight a new approach for controlling the behavior of spiking networks with realistic temporal dynamics, opening up the potential for using these networks to process spike patterns with complex temporal information.