Direct-to-Event Spiking Neural Network Transfer

TL;DR

This paper investigates how to convert directly coded spiking neural networks into event-based representations to improve energy efficiency without sacrificing performance.

Contribution

It provides the first systematic analysis and methods for transferring pretrained SNNs from direct coding to event-based computation.

Findings

Identifies key challenges in the transfer process.

Proposes methods to enable energy-efficient transfer.

Preserves model performance after transfer.

Abstract

Spiking Neural Networks (SNNs) have gained increasing attention due to their potential for low-power computation on neuromorphic hardware. A widely adopted training strategy for SNNs is direct coding, which enable backpropagation on neuron implementations using continuous-valued surrogate activations. However, recent studies have shown that direct-coded SNNs remain substantially less energy-efficient than their event-based counterparts, limiting their practical deployment in energy sensitive scenarios. Still, to promote the reusability of pretrained SNN database on direct code, this motivates an important yet underexplored question: How can a SNN pretrained with direct code be effectively converted into an event-based representation? In this research, we present the first systematic investigation into this transfer problem, analyze the key challenges that arise when transitioning from direct-coded to event-based computation and propose a set of methods to enable energy-efficient transfer while preserving model performance.