Opinion Dynamics-based Coalition Formation for Federated Learning in Heterogeneous IoT Systems

2026-09-17

AI Summary

本文针对联邦学习在异构物联网系统中的统计异质性问题，通过基于意见动态的联盟形成方法，在局部权重空间直接形成客户端联盟并进行聚合，有效提高了模型性能。

Abstract

Federated learning (FL) enables privacy-preserving, on-device training across heterogeneous Internet-of-Things (IoT) deployments such as smart-city water-metering networks, where each smart meter observes a household-specific consumption time series. Under such statistical heterogeneity, the standard Federated Averaging (FedAvg) aggregation averages dissimilar local models into a single global model that may fail to capture client-specific patterns. We address this by forming client coalitions directly in the local-weight space and aggregating at the coalition level. Extending a prior weight-driven coalition-formation scheme, we model coalition formation as a Hegselmann-Krause (HK) bounded-confidence opinion-dynamics process acting on the local weights, and develop variants of the HK interaction based on Euclidean-distance and cosine-similarity confidence criteria. The framework is applied to short-term water-consumption forecasting with local Long Short-Term Memory (LSTM) models and evaluated against FedAvg, Per-FedAvg, FedProx, and FedAvg with Euclidean-distance or cosine-similarity coalition formation. Experiments on a real smart-metering dataset of water consumption show that the proposed HK-based coalition formation produces stable, endogenous coalition structures within at most ten inner iterations, incurs no additional client-side computation or communication compared to FedAvg, and reduces the average MAE by up to 54% relative to FedAvg, 39% relative to FedProx, and 24% relative to Per-FedAvg, while achieving the highest global accuracy (83-85%).

Problem

• Federated Learning
• Statistical Heterogeneity
• Federated Averaging
• Client-specific Patterns
• IoT Systems

Innovation

• Hegselmann-Krause
• coalition formation
• federated learning
• opinion dynamics
• IoT systems