Fast On-Device Adaptation for Spiking Neural Networks via Online-Within-Online Meta-Learning

TL;DR

This paper introduces OWOML-SNN, an online meta-learning method for spiking neural networks that enables quick, lifelong adaptation on edge devices without backpropagation.

Contribution

It proposes a novel online-within-online meta-learning rule for SNNs that supports lifelong, backprop-free adaptation on neuromorphic hardware.

Findings

Enables rapid on-device adaptation with minimal data

Operates without backpropagation, suitable for neuromorphic hardware

Supports lifelong learning on streaming tasks

Abstract

Spiking Neural Networks (SNNs) have recently gained popularity as machine learning models for on-device edge intelligence for applications such as mobile healthcare management and natural language processing due to their low power profile. In such highly personalized use cases, it is important for the model to be able to adapt to the unique features of an individual with only a minimal amount of training data. Meta-learning has been proposed as a way to train models that are geared towards quick adaptation to new tasks. The few existing meta-learning solutions for SNNs operate offline and require some form of backpropagation that is incompatible with the current neuromorphic edge-devices. In this paper, we propose an online-within-online meta-learning rule for SNNs termed OWOML-SNN, that enables lifelong learning on a stream of tasks, and relies on local, backprop-free, nested updates.