# Adaptive Graph Pruning with Sudden-Events Evaluation for Traffic Prediction using Online Semi-Decentralized ST-GNNs Source: [Adaptive Graph Pruning with Sudden-Events Evaluation for Traffic Prediction using Online Semi-Decentralized ST-GNNs]() Collection index: [llms.txt]() Data updated: 2025-12-23T05:00:04.997Z Authors: Ivan Kralj, Lodovico Giaretta, Gordan Je\\v{z}i\\'c, Ivana Podnar \\v{Z}arko, \\v{S}ar\\=unas Girdzijauskas Published: 2025-12-19T08:48:36.000Z ## Abstract Spatio-Temporal Graph Neural Networks (ST-GNNs) are well-suited for processing high-frequency data streams from geographically distributed sensors in smart mobility systems. However, their deployment at the edge across distributed compute nodes (cloudlets) createssubstantial communication overhead due to repeated transmission of overlapping node features between neighbouring cloudlets. To address this, we propose an adaptive pruning algorithm that dynamically filters redundant neighbour features while preserving the most informative spatial context for prediction. The algorithm adjusts pruning rates based on recent model performance, allowing each cloudlet to focus on regions experiencing traffic changes without compromising accuracy. Additionally, we introduce the Sudden Event Prediction Accuracy (SEPA), a novel event-focused metric designed to measure responsiveness to traffic slowdowns and recoveries, which are often missed by standard error metrics. We evaluate our approach in an online semi-decentralized setting with traditional FL, server-free FL, and Gossip Learning on two large-scale traffic datasets, PeMS-BAY and PeMSD7-M, across short-, mid-, and long-term prediction horizons. Experiments show that, in contrast to standard metrics, SEPA exposes the true value of spatial connectivity in predicting dynamic and irregular traffic. Our adaptive pruning algorithm maintains prediction accuracy while significantly lowering communication cost in all online semi-decentralized settings, demonstrating that communication can be reduced without compromising responsiveness to critical traffic events. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Traffic Prediction on PeMS-BAY 15 min horizon (test)](): Traffic Prediction; PeMS-BAY 15 min horizon (test); MAE: 1.5. - [Traffic Prediction on PeMSD7-M 15 min horizon (test)](): Traffic Prediction; PeMSD7-M 15 min horizon (test); MAE: 2.44. - [Traffic Prediction on PeMS-BAY 60 min horizon (test)](): Traffic Prediction; PeMS-BAY 60 min horizon (test); MAE: 2.7. - [Traffic Prediction on PeMSD7-M (30 min horizon, test)](): Traffic Prediction; PeMSD7-M 30 min horizon (test); MAE: 3.37. - [Traffic Prediction on PeMSD7-M 60 min horizon (test)](): Traffic Prediction; PeMSD7-M 60 min horizon (test); MAE: 4.59. - [Traffic Prediction on PeMS-BAY 30 min horizon (test)](): Traffic Prediction; PeMS-BAY 30 min horizon (test); MAE: 2.04. ## Original sources and implementation [Original paper]() [PDF]()