Spiking Synaptic Penalty: Appropriate Penalty Term for Energy-Efficient Spiking Neural Networks

TL;DR

This paper introduces a novel penalty term for training spiking neural networks that directly optimizes energy consumption, effectively balancing energy efficiency and accuracy without altering network architecture.

Contribution

The paper proposes a new penalty term for SNN training that reduces energy consumption while maintaining accuracy, addressing the energy-accuracy trade-off.

Findings

Significant reduction in energy consumption during SNN training.

Maintains high accuracy comparable to traditional methods.

Effective in image classification tasks.

Abstract

Spiking neural networks (SNNs) are energy-efficient neural networks because of their spiking nature. However, as the spike firing rate of SNNs increases, the energy consumption does as well, and thus, the advantage of SNNs diminishes. Here, we tackle this problem by introducing a novel penalty term for the spiking activity into the objective function in the training phase. Our method is designed so as to optimize the energy consumption metric directly without modifying the network architecture. Therefore, the proposed method can reduce the energy consumption more than other methods while maintaining the accuracy. We conducted experiments for image classification tasks, and the results indicate the effectiveness of the proposed method, which mitigates the dilemma of the energy--accuracy trade-off.