FedDis: A Causal Disentanglement Framework for Federated Traffic Prediction

2026-01-30

AI Summary

This work addresses the tension between model generalization and local adaptability in federated traffic forecasting caused by non-IID data distributions. To resolve this, we propose FedDis, a causal disentanglement framework that, for the first time, introduces causal disentanglement into federated spatiotemporal prediction. FedDis employs a dual-branch architecture to separately model client-specific dynamics and cross-client shared spatiotemporal patterns, enforcing information orthogonality between the two through mutual information minimization. This design enables effective unification of global knowledge transfer and local personalization while preserving data privacy. Extensive experiments on four real-world traffic datasets demonstrate that FedDis significantly outperforms existing methods, achieving state-of-the-art performance in prediction accuracy, computational efficiency, and scalability.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Federated learning offers a promising paradigm for privacy-preserving traffic prediction, yet its performance is often challenged by the non-identically and independently distributed (non-IID) nature of decentralized traffic data. Existing federated methods frequently struggle with this data heterogeneity, typically entangling globally shared patterns with client-specific local dynamics within a single representation. In this work, we postulate that this heterogeneity stems from the entanglement of two distinct generative sources: client-specific localized dynamics and cross-client global spatial-temporal patterns. Motivated by this perspective, we introduce FedDis, a novel framework that, to the best of our knowledge, is the first to leverage causal disentanglement for federated spatial-temporal prediction. Architecturally, FedDis comprises a dual-branch design wherein a Personalized Bank learns to capture client-specific factors, while a Global Pattern Bank distills common knowledge. This separation enables robust cross-client knowledge transfer while preserving high adaptability to unique local environments. Crucially, a mutual information minimization objective is employed to enforce informational orthogonality between the two branches, thereby ensuring effective disentanglement. Comprehensive experiments conducted on four real-world benchmark datasets demonstrate that FedDis consistently achieves state-of-the-art performance, promising efficiency, and superior expandability.

Problem

• federated learning
• traffic prediction
• non-IID
• data heterogeneity
• causal disentanglement

Innovation

• causal disentanglement
• federated learning
• spatial-temporal prediction
• non-IID data
• mutual information minimization