Decoupled Dynamic Spatial-Temporal Graph Neural Network for Traffic Forecasting

TL;DR

This paper introduces a novel decoupled framework for traffic forecasting using spatial-temporal graph neural networks, effectively separating diffusion and inherent signals to improve prediction accuracy on real-world datasets.

Contribution

The paper proposes a decoupled spatial-temporal framework that separates diffusion and inherent traffic signals, enhancing modeling performance over existing methods.

Findings

Outperforms state-of-the-art models on four real-world datasets.

Effectively captures dynamic characteristics of traffic networks.

Improves traffic forecasting accuracy.

Abstract

We all depend on mobility, and vehicular transportation affects the daily lives of most of us. Thus, the ability to forecast the state of traffic in a road network is an important functionality and a challenging task. Traffic data is often obtained from sensors deployed in a road network. Recent proposals on spatial-temporal graph neural networks have achieved great progress at modeling complex spatial-temporal correlations in traffic data, by modeling traffic data as a diffusion process. However, intuitively, traffic data encompasses two different kinds of hidden time series signals, namely the diffusion signals and inherent signals. Unfortunately, nearly all previous works coarsely consider traffic signals entirely as the outcome of the diffusion, while neglecting the inherent signals, which impacts model performance negatively. To improve modeling performance, we propose a novel Decoupled Spatial-Temporal Framework (DSTF) that separates the diffusion and inherent traffic information in a data-driven manner, which encompasses a unique estimation gate and a residual decomposition mechanism. The separated signals can be handled subsequently by the diffusion and inherent modules separately. Further, we propose an instantiation of DSTF, Decoupled Dynamic Spatial-Temporal Graph Neural Network (D2STGNN), that captures spatial-temporal correlations and also features a dynamic graph learning module that targets the learning of the dynamic characteristics of traffic networks. Extensive experiments with four real-world traffic datasets demonstrate that the framework is capable of advancing the state-of-the-art.