Training Stronger Spiking Neural Networks with Biomimetic Adaptive Internal Association Neurons

TL;DR

This paper introduces a novel Adaptive Internal Association neuron model for spiking neural networks, inspired by biological associative phenomena, leading to improved performance and efficiency on neuromorphic datasets.

Contribution

The paper proposes the AIA neuron model that incorporates internal associative learning within neurons, a novel approach in SNNs inspired by biological ALTP phenomena.

Findings

Achieves state-of-the-art accuracy on DVS-CIFAR10 and N-CARS datasets.

Enhances spike specificity and reduces spike count.

No additional parameters required at inference.

Abstract

As the third generation of neural networks, spiking neural networks (SNNs) are dedicated to exploring more insightful neural mechanisms to achieve near-biological intelligence. Intuitively, biomimetic mechanisms are crucial to understanding and improving SNNs. For example, the associative long-term potentiation (ALTP) phenomenon suggests that in addition to learning mechanisms between neurons, there are associative effects within neurons. However, most existing methods only focus on the former and lack exploration of the internal association effects. In this paper, we propose a novel Adaptive Internal Association~(AIA) neuron model to establish previously ignored influences within neurons. Consistent with the ALTP phenomenon, the AIA neuron model is adaptive to input stimuli, and internal associative learning occurs only when both dendrites are stimulated at the same time. In addition, we employ weighted weights to measure internal associations and introduce intermediate caches to reduce the volatility of associations. Extensive experiments on prevailing neuromorphic datasets show that the proposed method can potentiate or depress the firing of spikes more specifically, resulting in better performance with fewer spikes. It is worth noting that without adding any parameters at inference, the AIA model achieves state-of-the-art performance on DVS-CIFAR10~(83.9\%) and N-CARS~(95.64\%) datasets.