SiLIF: Structured State Space Model Dynamics and Parametrization for Spiking Neural Networks

TL;DR

This paper introduces two novel SSM-inspired spiking neuron models, SiLIF, which achieve state-of-the-art results in speech recognition tasks while maintaining efficiency, inspired by stable state space model dynamics.

Contribution

The paper proposes two new SSM-inspired Leaky Integrate-and-Fire neuron models that improve performance and stability in spiking neural networks.

Findings

Achieved state-of-the-art performance on speech recognition datasets.

Demonstrated a better performance-efficiency trade-off than traditional SSMs.

Surpassed SSMs in efficiency using synaptic delays.

Abstract

Multi-state spiking neurons combine sparse binary activations with rich second-order nonlinear recurrent dynamics, making them a promising alternative to standard deep learning models. However, gradient propagation through these dynamics often leads to instabilities that hinder scalability and performance. Inspired by the stable training and strong performance of state space models (SSMs) on long sequences, we introduce two SSM-inspired Leaky Integrate-and-Fire (SiLIF) neuron models. The first extends a two-state neuron with a learnable discretization timestep and logarithmic reparametrization, while the second additionally incorporates the initialization scheme and structure of complex-state SSMs, enabling oscillatory regimes. Our two SiLIF models achieve new state-of-the-art performance among spiking neuron models on both event-based and raw-audio speech recognition datasets. We further demonstrate a favorable performance-efficiency trade-off compared to SSMs, even surpassing them while using half the computational cost through the use of synaptic delays. Our code is available at https://github.com/Maxtimer97/SSM-inspired-LIF.