Energy-efficient Knowledge Distillation for Spiking Neural Networks

TL;DR

This paper introduces a novel knowledge distillation method for spiking neural networks that improves accuracy while significantly reducing energy consumption by decreasing spike activity.

Contribution

It proposes a heterogeneous temperature-based knowledge distillation technique tailored for SNNs to enhance energy efficiency alongside accuracy.

Findings

Achieves up to 0.09% higher accuracy on MNIST.

Reduces spike activity by 65% compared to conventional methods.

Demonstrates effectiveness over other SNN compression techniques.

Abstract

Spiking neural networks (SNNs) have been gaining interest as energy-efficient alternatives of conventional artificial neural networks (ANNs) due to their event-driven computation. Considering the future deployment of SNN models to constrained neuromorphic devices, many studies have applied techniques originally used for ANN model compression, such as network quantization, pruning, and knowledge distillation, to SNNs. Among them, existing works on knowledge distillation reported accuracy improvements of student SNN model. However, analysis on energy efficiency, which is also an important feature of SNN, was absent. In this paper, we thoroughly analyze the performance of the distilled SNN model in terms of accuracy and energy efficiency. In the process, we observe a substantial increase in the number of spikes, leading to energy inefficiency, when using the conventional knowledge distillation methods. Based on this analysis, to achieve energy efficiency, we propose a novel knowledge distillation method with heterogeneous temperature parameters. We evaluate our method on two different datasets and show that the resulting SNN student satisfies both accuracy improvement and reduction of the number of spikes. On MNIST dataset, our proposed student SNN achieves up to 0.09% higher accuracy and produces 65% less spikes compared to the student SNN trained with conventional knowledge distillation method. We also compare the results with other SNN compression techniques and training methods.