SpikySpace: A Spiking State Space Model for Energy-Efficient Time Series Forecasting

TL;DR

SpikySpace introduces a novel spiking state-space model that significantly improves energy efficiency and accuracy in real-time multivariate time series forecasting on edge devices.

Contribution

It is the first fully spiking state-space model that reduces computational complexity and energy consumption while maintaining high accuracy in time series prediction.

Findings

Outperforms existing SNN models by up to 3.0% in accuracy.

Reduces energy consumption by over 96.1%.

Achieves linear time complexity in attention computation.

Abstract

Time-series forecasting in domains like traffic management and industrial monitoring often requires real-time, energy-efficient processing on edge devices with limited resources. Spiking neural networks (SNNs) offer event-driven computation and ultra-low power and have been proposed for use in this space. Unfortunately, existing SNN-based time-series forecasters often use complex transformer blocks. To address this issue, we propose SpikySpace, a spiking state-space model (SSM) that reduces the quadratic cost in the attention block to linear time via spiking selective scanning. Further, we introduce PTsoftplus and PTSiLU, two efficient approximations of SiLU and Softplus that replace costly exponential and division operations with simple bit-shifts. Evaluated on four multivariate time-series benchmarks, SpikySpace outperforms the leading SNN in terms of accuracy by up to 3.0% while reducing energy consumption by over 96.1%. As the first fully spiking state-space model, SpikySpace bridges neuromorphic efficiency with modern sequence modeling, opening a practical path toward efficient time series forecasting systems. Our code is available at https://anonymous.4open.science/r/SpikySpace.