Biologically Inspired Dynamic Thresholds for Spiking Neural Networks

TL;DR

This paper introduces a novel bioinspired dynamic threshold scheme for spiking neural networks that improves robustness and performance in complex real-world tasks by mimicking biological neuron regulation mechanisms.

Contribution

The paper proposes the BDETT scheme, a biologically inspired dynamic threshold for SNNs, bridging the gap between neuroscience and machine learning.

Findings

BDETT outperforms static and heuristic thresholds in various tasks.

The scheme enhances robustness under noisy and dynamic conditions.

It maintains neuronal homeostasis in complex environments.

Abstract

The dynamic membrane potential threshold, as one of the essential properties of a biological neuron, is a spontaneous regulation mechanism that maintains neuronal homeostasis, i.e., the constant overall spiking firing rate of a neuron. As such, the neuron firing rate is regulated by a dynamic spiking threshold, which has been extensively studied in biology. Existing work in the machine learning community does not employ bioinspired spiking threshold schemes. This work aims at bridging this gap by introducing a novel bioinspired dynamic energy-temporal threshold (BDETT) scheme for spiking neural networks (SNNs). The proposed BDETT scheme mirrors two bioplausible observations: a dynamic threshold has 1) a positive correlation with the average membrane potential and 2) a negative correlation with the preceding rate of depolarization. We validate the effectiveness of the proposed BDETT on robot obstacle avoidance and continuous control tasks under both normal conditions and various degraded conditions, including noisy observations, weights, and dynamic environments. We find that the BDETT outperforms existing static and heuristic threshold approaches by significant margins in all tested conditions, and we confirm that the proposed bioinspired dynamic threshold scheme offers homeostasis to SNNs in complex real-world tasks.