Federated learning improves cross-region traffic flow prediction accuracy
FedeRICo: Federated Region-Influenced Coupling for Traffic Flow Prediction
Artificial IntelligenceDistributed, Parallel, and Cluster Computing
Summary
Traffic forecasts help cities manage congestion but are hard to do when data is spread across different regions and can't be shared due to privacy or business reasons. The authors present a new method called FedeRICo that lets different areas collaborate on traffic forecasting without sharing raw data. Their approach carefully combines shared global patterns with local adjustments and exchanges limited border information to capture how nearby regions influence each other. Tests on real-world data show their method predicts traffic better than existing federated approaches while running efficiently.
What this means in practice
- •For urban traffic managers: Improve traffic flow forecasts across administrative boundaries without sharing sensitive raw data to better manage congestion.
- •For smart city system developers: Develop federated traffic prediction systems that combine shared global models with local corrections to handle heterogeneous sensor networks.
Authors
Fermin Orozco, Man Luo, Johan Wahlström
Abstract
Urban traffic forecasting often relies on information distributed across stakeholders who may be unable to share raw data due to privacy or commercial constraints, motivating federated spatial-temporal approaches. In such federated settings, each client observes traffic over a distinct sensor subgraph with its own spatial topology and temporal dynamics, leading to significant heterogeneity across clients. Existing federated spatial-temporal methods typically rely on model parameter aggregation and provide limited mechanisms for recovering spatial dependencies across client boundaries. This introduces two key limitations. Specifically, parameter aggregation across heterogeneous graph domains tends to dilute client-specific representations, while road network partitioning breaks the propagation of traffic dynamics across client boundaries. To address these challenges, we propose FedeRICo, a federated traffic forecasting framework that combines gradient-level collaboration with boundary-aware residual communication. FedeRICo employs a dual-branch forecasting architecture in which a globally guided branch captures transferable forecasting structure, while a private residual branch preserves client-specific corrections and incorporates boundary residual signals. The global branch is coordinated through gradient alignment across all clients, enabling collaborative optimisation without destructive parameter interference. To recover cross-client spatial dependencies, boundary messages are extracted through a trend-residual decomposition that suppresses periodic structure and communicates only transient spatial-temporal residual signals between physically adjacent clients. Experiments across four real-world traffic forecasting benchmarks demonstrate that FedeRICo consistently outperforms state-of-the-art federated spatial-temporal baselines while maintaining competitive training runtime.