Federated Learning of Spiking Neural Networks under Heterogeneous Temporal Resolutions

2026-05-14

AI Summary

This work addresses the challenge of effectively aggregating spiking neural network (SNN) models in federated learning when edge devices exhibit heterogeneous temporal resolutions in their time-series data. To overcome the performance degradation caused by temporal misalignment under conventional federated averaging, the paper proposes the first federated learning framework tailored for stateful-neuron SNNs under heterogeneous time resolutions. The approach introduces a time-adaptive mechanism for neuron parameter alignment and model aggregation, enabling each client to train locally at its native temporal scale while remaining compatible with the global model. Experimental results on the SHD and DVS-Gesture datasets demonstrate that the proposed framework substantially recovers accuracy lost due to temporal resolution discrepancies, achieving efficient and effective federated training across heterogeneous edge devices.

Abstract

Spiking neural networks (SNNs) are biologically inspired energy-efficient models that use sparse binary spike-based communication between neurons, making them attractive for resource-constrained edge devices. Federated learning enables such devices to train collaboratively without sharing raw data. In time-series applications, edge devices often collect data at different time resolutions due to hardware and energy constraints. This temporal heterogeneity poses a fundamental challenge for federated learning: parameters learned at one temporal resolution do not necessarily transfer directly to another, which might result in the naive federated averaging being ineffective. Targeting SNNs and, more broadly, deep networks with stateful neurons, we propose a federated learning framework that addresses this temporal resolution mismatch. We investigate how neuron parameters learned from data at different temporal resolutions and model aggregation should be integrated. We evaluate the proposed framework across two SNN-native benchmark datasets (SHD and DVS-Gesture) under a range of resolution heterogeneity scenarios. Our results show that the proposed adaptation methods can substantially recover accuracy lost due to temporal mismatch, hence enabling each client to train at their local temporal resolution while remaining compatible with the global model.

Problem

• Federated Learning
• Spiking Neural Networks
• Temporal Heterogeneity
• Time Resolution Mismatch
• Edge Devices

Innovation

• Federated Learning
• Spiking Neural Networks
• Temporal Heterogeneity
• Time Resolution Mismatch
• Model Aggregation