Gated Residual Recurrent Graph Neural Networks for Traffic Prediction

TLDR

Traffic prediction is crucial for traffic management and public safety, yet remains difficult due to complex spatial dependencies, temporal dynamics, and the limitations of existing CNN–RNN combinations that fail to capture the connectivity and globality of road networks. This study introduces a residual recurrent graph neural network (Res‑RGNN) and a hop scheme to jointly model graph‑based spatial dependencies and periodic temporal dynamics, culminating in an end‑to‑end multiple Res‑RGNN (MRes‑RGNN) framework for traffic prediction. The MRes‑RGNN framework employs residual recurrent GNN layers to mitigate gradient vanishing, incorporates a hop‑based temporal dependency module, and integrates multiple such networks in an end‑to‑end architecture to capture both spatial and temporal patterns. Experiments on two traffic datasets demonstrate that MRes‑RGNN significantly outperforms state‑of‑the‑art methods.

Abstract

Traffic prediction is of great importance to traffic management and public safety, and very challenging as it is affected by many complex factors, such as spatial dependency of complicated road networks and temporal dynamics, and many more. The factors make traffic prediction a challenging task due to the uncertainty and complexity of traffic states. In the literature, many research works have applied deep learning methods on traffic prediction problems combining convolutional neural networks (CNNs) with recurrent neural networks (RNNs), which CNNs are utilized for spatial dependency and RNNs for temporal dynamics. However, such combinations cannot capture the connectivity and globality of traffic networks. In this paper, we first propose to adopt residual recurrent graph neural networks (Res-RGNN) that can capture graph-based spatial dependencies and temporal dynamics jointly. Due to gradient vanishing, RNNs are hard to capture periodic temporal correlations. Hence, we further propose a novel hop scheme into Res-RGNN to utilize the periodic temporal dependencies. Based on Res-RGNN and hop Res-RGNN, we finally propose a novel end-to-end multiple Res-RGNNs framework, referred to as “MRes-RGNN”, for traffic prediction. Experimental results on two traffic datasets have demonstrated that the proposed MRes-RGNN outperforms state-of-the-art methods significantly.