Traffic Flow Forecasting with Spatial-Temporal Graph Diffusion Network

TL;DR

This paper introduces ST-GDN, a hierarchical graph neural network that captures both local and global spatial dependencies and multi-scale temporal dynamics for more accurate citywide traffic flow forecasting.

Contribution

The paper proposes a novel spatial-temporal graph diffusion network that models global inter-region dependencies and multi-resolution temporal patterns, improving traffic prediction accuracy.

Findings

ST-GDN outperforms state-of-the-art baselines on real traffic datasets.

The model effectively captures global spatial dependencies.

Multi-scale attention enhances temporal dynamics modeling.

Abstract

Accurate forecasting of citywide traffic flow has been playing critical role in a variety of spatial-temporal mining applications, such as intelligent traffic control and public risk assessment. While previous work has made significant efforts to learn traffic temporal dynamics and spatial dependencies, two key limitations exist in current models. First, only the neighboring spatial correlations among adjacent regions are considered in most existing methods, and the global inter-region dependency is ignored. Additionally, these methods fail to encode the complex traffic transition regularities exhibited with time-dependent and multi-resolution in nature. To tackle these challenges, we develop a new traffic prediction framework-Spatial-Temporal Graph Diffusion Network (ST-GDN). In particular, ST-GDN is a hierarchically structured graph neural architecture which learns not only the local region-wise geographical dependencies, but also the spatial semantics from a global perspective. Furthermore, a multi-scale attention network is developed to empower ST-GDN with the capability of capturing multi-level temporal dynamics. Experiments on several real-life traffic datasets demonstrate that ST-GDN outperforms different types of state-of-the-art baselines. Source codes of implementations are available at https://github.com/jill001/ST-GDN.