Not All Timesteps Matter Equally: Selective Alignment Knowledge Distillation for Spiking Neural Networks

TL;DR

This paper introduces SeAl-KD, a novel knowledge distillation method for spiking neural networks that selectively aligns knowledge at important timesteps, improving performance over existing methods.

Contribution

The paper proposes a selective alignment approach for knowledge distillation in SNNs, addressing the issue of uniform treatment of all timesteps and enhancing distillation effectiveness.

Findings

SeAl-KD improves SNN performance on various datasets.

Selective alignment reduces erroneous timestep influence.

The method outperforms existing distillation techniques.

Abstract

Spiking neural networks (SNNs), which are brain-inspired and spike-driven, achieve high energy efficiency. However, a performance gap between SNNs and artificial neural networks (ANNs) still remains. Knowledge distillation (KD) is commonly adopted to improve SNN performance, but existing methods typically enforce uniform alignment across all timesteps, either from a teacher network or through inter-temporal self-distillation, implicitly assuming that per-timestep predictions should be treated equally. In practice, SNN predictions vary and evolve over time, and intermediate timesteps need not all be individually correct even when the final aggregated output is correct. Under such conditions, effective distillation should not force every timestep toward the same supervision target, but instead provide corrective guidance to erroneous timesteps while preserving useful temporal dynamics. To address this issue, we propose Selective Alignment Knowledge Distillation (SeAl-KD), which selectively aligns class-level and temporal knowledge by equalizing competing logits at erroneous timesteps and reweighting temporal alignment based on confidence and inter-timestep similarity. Extensive experiments on static image and neuromorphic event-based datasets demonstrate consistent improvements over existing distillation methods. The code is available at https://github.com/KaiSUN1/SeAl