Federated Learning with Graph-Based Aggregation for Traffic Forecasting

2025-07-13

AI Summary

In traffic forecasting, conventional federated learning (FL) struggles to capture cross-regional spatial dependencies while incurring high computational overhead. To address this, we propose a lightweight graph-aware federated learning framework. Our method integrates neighborhood aggregation—inspired by graph neural networks (GNNs)—into client-side parameter updates, leveraging a predefined road network topology to guide weighted model aggregation. This relaxes FedAvg’s strong independence assumption among clients and explicitly models spatial correlations without compromising data privacy. Crucially, our approach avoids full GNN training, eliminating costly global graph computations and significantly reducing both communication and computational complexity. Extensive experiments on METR-LA and PEMS-BAY demonstrate that our method achieves performance on par with state-of-the-art centralized models and recent graph-based federated baselines, while offering superior efficiency and scalability.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Planning, Routing, and Scheduling: Planning with Language Models
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• 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

Abstract

In traffic prediction, the goal is to estimate traffic speed or flow in specific regions or road segments using historical data collected by devices deployed in each area. Each region or road segment can be viewed as an individual client that measures local traffic flow, making Federated Learning (FL) a suitable approach for collaboratively training models without sharing raw data. In centralized FL, a central server collects and aggregates model updates from multiple clients to build a shared model while preserving each client's data privacy. Standard FL methods, such as Federated Averaging (FedAvg), assume that clients are independent, which can limit performance in traffic prediction tasks where spatial relationships between clients are important. Federated Graph Learning methods can capture these dependencies during server-side aggregation, but they often introduce significant computational overhead. In this paper, we propose a lightweight graph-aware FL approach that blends the simplicity of FedAvg with key ideas from graph learning. Rather than training full models, our method applies basic neighbourhood aggregation principles to guide parameter updates, weighting client models based on graph connectivity. This approach captures spatial relationships effectively while remaining computationally efficient. We evaluate our method on two benchmark traffic datasets, METR-LA and PEMS-BAY, and show that it achieves competitive performance compared to standard baselines and recent graph-based federated learning techniques.

Problem

• Traffic prediction using Federated Learning with spatial dependencies
• Standard FL ignores spatial relationships in traffic data
• Lightweight graph-aware FL balances efficiency and performance

Innovation

• Lightweight graph-aware Federated Learning approach
• Basic neighbourhood aggregation for parameter updates
• Weighting client models based on graph connectivity