Fast Exploration of the Impact of Precision Reduction on Spiking Neural Networks

TL;DR

This paper presents a method using Interval Arithmetic to efficiently explore how precision reduction impacts Spiking Neural Networks, enabling faster optimization and potential network size reduction for edge computing.

Contribution

It introduces an IA-based exploration methodology for precision reduction in SNNs, improving efficiency and enabling finer-grained analysis compared to traditional methods.

Findings

Significantly reduced exploration time.

Ability to further minimize network parameters.

Enhanced precision-error detection for SNNs.

Abstract

Approximate Computing (AxC) techniques trade off the computation accuracy for performance, energy, and area reduction gains. The trade-off is particularly convenient when the applications are intrinsically tolerant to some accuracy loss, as in the Spiking Neural Networks (SNNs) case. SNNs are a practical choice when the target hardware reaches the edge of computing, but this requires some area minimization strategies. In this work, we employ an Interval Arithmetic (IA) model to develop an exploration methodology that takes advantage of the capability of such a model to propagate the approximation error to detect when the approximation exceeds tolerable limits by the application. Experimental results confirm the capability of reducing the exploration time significantly, providing the chance to reduce the network parameters' size further and with more fine-grained results.