Event-based backpropagation on the neuromorphic platform SpiNNaker2

TL;DR

This paper presents the first implementation of event-based backpropagation on SpiNNaker2 neuromorphic hardware, enabling energy-efficient, sparse training of spiking neural networks with demonstrated proof-of-concept results.

Contribution

It introduces an event-based backpropagation algorithm adapted for SpiNNaker2, facilitating on-chip training of SNNs with exact gradient computation using sparse error communication.

Findings

First implementation of event-based backpropagation on SpiNNaker2

Successful proof-of-concept with batch-parallelized on-chip training

Efficient hybrid training methods demonstrated

Abstract

Neuromorphic computing aims to replicate the brain's capabilities for energy efficient and parallel information processing, promising a solution to the increasing demand for faster and more efficient computational systems. Efficient training of neural networks on neuromorphic hardware requires the development of training algorithms that retain the sparsity of spike-based communication during training. Here, we report on the first implementation of event-based backpropagation on the SpiNNaker2 neuromorphic hardware platform. We use EventProp, an algorithm for event-based backpropagation in spiking neural networks (SNNs), to compute exact gradients using sparse communication of error signals between neurons. Our implementation computes multi-layer networks of leaky integrate-and-fire neurons using discretized versions of the differential equations and their adjoints, and uses event packets to transmit spikes and error signals between network layers. We demonstrate a proof-of-concept of batch-parallelized, on-chip training of SNNs using the Yin Yang dataset, and provide an off-chip implementation for efficient prototyping, hyper-parameter search, and hybrid training methods.