Unlocking Dynamic Inter-Client Spatial Dependencies: A Federated Spatio-Temporal Graph Learning Method for Traffic Flow Forecasting

TL;DR

This paper introduces FedSTGD, a federated learning framework that models and reconstructs dynamic inter-client spatial dependencies in traffic forecasting, improving accuracy while respecting data privacy constraints.

Contribution

FedSTGD is the first framework to effectively model dynamic inter-client spatial dependencies in federated traffic forecasting, combining nonlinear computation decomposition and node embedding augmentation.

Findings

FedSTGD outperforms state-of-the-art baselines in traffic prediction metrics.

The modules in FedSTGD significantly improve modeling of dynamic dependencies.

FedSTGD demonstrates robustness to hyperparameter variations.

Abstract

Spatio-temporal graphs are powerful tools for modeling complex dependencies in traffic time series. However, the distributed nature of real-world traffic data across multiple stakeholders poses significant challenges in modeling and reconstructing inter-client spatial dependencies while adhering to data locality constraints. Existing methods primarily address static dependencies, overlooking their dynamic nature and resulting in suboptimal performance. In response, we propose Federated Spatio-Temporal Graph with Dynamic Inter-Client Dependencies (FedSTGD), a framework designed to model and reconstruct dynamic inter-client spatial dependencies in federated learning. FedSTGD incorporates a federated nonlinear computation decomposition module to approximate complex graph operations. This is complemented by a graph node embedding augmentation module, which alleviates performance degradation arising from the decomposition. These modules are coordinated through a client-server collective learning protocol, which decomposes dynamic inter-client spatial dependency learning tasks into lightweight, parallelizable subtasks. Extensive experiments on four real-world datasets demonstrate that FedSTGD achieves superior performance over state-of-the-art baselines in terms of RMSE, MAE, and MAPE, approaching that of centralized baselines. Ablation studies confirm the contribution of each module in addressing dynamic inter-client spatial dependencies, while sensitivity analysis highlights the robustness of FedSTGD to variations in hyperparameters.