General Self-Prediction Enhancement for Spiking Neurons

TL;DR

This paper introduces a biologically inspired self-prediction mechanism for spiking neurons that improves training stability, accuracy, and broad applicability of Spiking Neural Networks by leveraging internal predictive currents.

Contribution

It proposes a novel self-prediction enhanced neuron model that creates a continuous gradient path and aligns with biological principles, addressing key training challenges in SNNs.

Findings

Consistent performance improvements across various architectures and tasks.

Enhanced training stability and accuracy in SNNs.

Broad applicability demonstrated through diverse experiments.

Abstract

Spiking Neural Networks (SNNs) are highly energy-efficient due to event-driven, sparse computation, but their training is challenged by spike non-differentiability and trade-offs among performance, efficiency, and biological plausibility. Crucially, mainstream SNNs ignore predictive coding, a core cortical mechanism where the brain predicts inputs and encodes errors for efficient perception. Inspired by this, we propose a self-prediction enhanced spiking neuron method that generates an internal prediction current from its input-output history to modulate membrane potential. This design offers dual advantages, it creates a continuous gradient path that alleviates vanishing gradients and boosts training stability and accuracy, while also aligning with biological principles, which resembles distal dendritic modulation and error-driven synaptic plasticity. Experiments show consistent performance gains across diverse architectures, neuron types, time steps, and tasks demonstrating broad applicability for enhancing SNNs.