TESS: A Scalable Temporally and Spatially Local Learning Rule for Spiking Neural Networks

TL;DR

TESS is a biologically inspired, scalable, and energy-efficient learning rule for spiking neural networks that achieves performance comparable to backpropagation while significantly reducing computational and memory overheads, enabling on-device edge learning.

Contribution

Introduces TESS, a local learning rule for SNNs that addresses temporal and spatial credit assignment with linear scaling, matching BPTT performance on challenging vision tasks.

Findings

TESS achieves comparable accuracy to BPTT on vision datasets.

TESS's complexity scales linearly with the number of neurons.

Enables efficient on-device learning for edge applications.

Abstract

The demand for low-power inference and training of deep neural networks (DNNs) on edge devices has intensified the need for algorithms that are both scalable and energy-efficient. While spiking neural networks (SNNs) allow for efficient inference by processing complex spatio-temporal dynamics in an event-driven fashion, training them on resource-constrained devices remains challenging due to the high computational and memory demands of conventional error backpropagation (BP)-based approaches. In this work, we draw inspiration from biological mechanisms such as eligibility traces, spike-timing-dependent plasticity, and neural activity synchronization to introduce TESS, a temporally and spatially local learning rule for training SNNs. Our approach addresses both temporal and spatial credit assignments by relying solely on locally available signals within each neuron, thereby allowing computational and memory overheads to scale linearly with the number of neurons, independently of the number of time steps. Despite relying on local mechanisms, we demonstrate performance comparable to the backpropagation through time (BPTT) algorithm, within $\sim1.4$ accuracy points on challenging computer vision scenarios relevant at the edge, such as the IBM DVS Gesture dataset, CIFAR10-DVS, and temporal versions of CIFAR10, and CIFAR100. Being able to produce comparable performance to BPTT while keeping low time and memory complexity, TESS enables efficient and scalable on-device learning at the edge.