SpikF-GO: Spiking Fourier Graph Operators for Multivariate Time Series Forecasting

2026-06-11

AI Summary

This work addresses the challenge that existing spiking neural networks (SNNs) struggle to explicitly model inter-variable dependencies in multivariate time series forecasting. To overcome this limitation, the study introduces graph structures into SNNs for the first time, proposing a novel approach based on hyper-variable graphs and spike-driven spectral processing. The method represents scalar observations as graph nodes and performs event-driven computation in the Fourier domain, incorporating learnable sparse frequency selection via Hard Concrete gating, complex-valued leaky integrate-and-fire (LIF) neurons, and central pattern generator-inspired positional encoding. Evaluated on eight benchmark datasets, the proposed model outperforms all existing SNN methods on average, surpasses the ANN-based FourierGNN baseline, and remains competitive even with substantially lower embedding dimensions—thereby achieving significantly reduced energy consumption.

Abstract

Spiking Neural Networks (SNNs) have emerged as an energy-efficient alternative to conventional neural networks, demonstrating strong performance in computer vision and robotics. More recently, SNNs have been applied to time series forecasting (TSF), with methods exploring spiking temporal backbones, spike-compatible positional encodings, Fourier-domain processing, and redesigned neuron dynamics. However, existing SNN forecasting approaches process variables independently, lacking explicit mechanisms for modeling inter-variable dependencies. This is a critical limitation in multivariate settings, where cross-variable correlations carry substantial predictive information. We propose Spiking Fourier Graph Operators (SpikF-GO), which addresses this gap by combining a hypervariate graph formulation in which every scalar observation becomes a graph node with spike-driven spectral processing. SpikF-GO introduces a Hard Concrete frequency gate for learnable sparse frequency selection and a Complex LIF gate that applies independent spiking neurons to real and imaginary Fourier components, preserving binary, event-driven computation throughout the spectral domain. We further present a variant incorporating Central Pattern Generator-based positional encodings for stronger long-range temporal modeling. Evaluated on eight benchmarks under a unified experimental protocol, SpikF-GO achieves the best average rank among all SNN methods and outperforms its ANN counterpart, FourierGNN, at reduced energy cost. SpikF-GO maintains competitive accuracy even at substantially smaller embedding dimensions, thereby achieving significant energy reductions. To our knowledge, this is among the first works to bring graph-based multivariate modeling into the spiking domain for TSF and the first to provide a unified comparison across SNN forecasting architectures under a common experimental protocol.

Problem

• Spiking Neural Networks
• Multivariate Time Series Forecasting
• Inter-variable Dependencies
• Graph-based Modeling
• Spectral Processing

Innovation

• Spiking Neural Networks
• Graph Operators
• Fourier Domain Processing
• Multivariate Time Series Forecasting
• Energy-Efficient AI