Small Graph Is All You Need: DeepStateGNN for Scalable Traffic Forecasting

TL;DR

The paper addresses scalable spatiotemporal traffic forecasting and reconstruction in the presence of missing sensors. It introduces DeepStateGNN, which replaces per-sensor graph nodes with a fixed-size Deep State Graph (DSG) composed of Deep State Nodes (DSNs) formed from spatial, semantic, environmental, and temporal similarities to capture latent traffic states. Key innovations include static/dynamic DSN assignments, a Long-Short Laplacian to fuse short- and long-range dependencies, DSN-state updates from embedded observations, and a multi-view inference pipeline that predicts $\hat{Y}_{traffic} \in \mathbb{R}^{|Q| \times H \times 2}$ for query locations. Evaluations on the METR-LA+ dataset show state-of-the-art performance in both forecasting and reconstruction, with superior scalability and robustness to sensor outages, highlighting practical benefits for real-world traffic monitoring systems.

Abstract

We propose a novel Graph Neural Network (GNN) model, named DeepStateGNN, for analyzing traffic data, demonstrating its efficacy in two critical tasks: forecasting and reconstruction. Unlike typical GNN methods that treat each traffic sensor as an individual graph node, DeepStateGNN clusters sensors into higher-level graph nodes, dubbed Deep State Nodes, based on various similarity criteria, resulting in a fixed number of nodes in a Deep State graph. The term "Deep State" nodes is a play on words, referencing hidden networks of power that, like these nodes, secretly govern traffic independently of visible sensors. These Deep State Nodes are defined by several similarity factors, including spatial proximity (e.g., sensors located nearby in the road network), functional similarity (e.g., sensors on similar types of freeways), and behavioral similarity under specific conditions (e.g., traffic behavior during rain). This clustering approach allows for dynamic and adaptive node grouping, as sensors can belong to multiple clusters and clusters may evolve over time. Our experimental results show that DeepStateGNN offers superior scalability and faster training, while also delivering more accurate results than competitors. It effectively handles large-scale sensor networks, outperforming other methods in both traffic forecasting and reconstruction accuracy.

Figures (2)

• Figure 1: Overall DeepStateGNN Architecture.
• Figure 2: Training time per epoch with a constant batch size.