S³ KD: Semantic-aware Spiking Neural Networks Self-distillation for Human Action Recognition

Human Action Recognition (HAR) using wearable sensors has become a topic of great interest in research due to its wide range of applications. However, achieving performance improvements in HAR often involves increasing model sizes, hindering the feasibility of deployment on resource-constrained wearable devices. Spiking Neural Networks (SNNs), inspired by biological neural networks, offer a promising alternative by leveraging the spiking nature of brain neurons and eventdriven activations. SNNs can thus reduce training complexity and power consumption significantly. Nevertheless, SNNs usually face accuracy degradation and require a substantial number of time steps to attain competitive performance when compared to conventional artificial neural networks (ANNs). To address these challenges, we propose a novel Semantic-aware Spiking neural network Self-distillation (S<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup>KD) framework to improve the accuracy of wearable-based HAR. First, we replace all activation functions with the Leaky-Integrate-and-Fire (LIF) neurons to construct SNN models. Next, we design a semantic-aware MixUp technique to construct more reliable soft labels for the S<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup>KD process. Furthermore, we introduce a knowledge preview scheme aimed at bolstering the transferability of model knowledge within SNNs. We validate the effectiveness of S<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup>KD through extensive experiments on three HAR datasets. The results indicate its superiority over previous SNN-based approaches, achieving competitive performance enhancements.