STAHGNet: Modeling Hybrid-grained Heterogenous Dependency Efficiently for Traffic Prediction

TL;DR

Traffic flow prediction is challenged by evolving hybrid-spatio-temporal heterogeneity. STAHGNet couples a Spatio-Temporal Aware Hybrid Graph Network with a Hybrid Graph Attention Module and a Coarse-grained Temporal Graph generator, augmented by domain-knowledge feature engineering and random neighbor sampling to capture both fine-grained and global temporal dependencies efficiently. The approach achieves state-of-the-art results on four public PeMS datasets, with notable improvements in MAE and significant memory savings, and is validated through extensive ablations and visualizations. This work offers a practical, scalable framework for intelligent transportation systems by modeling dynamic dependencies across multiple temporal granularities.

Abstract

Traffic flow prediction plays a critical role in the intelligent transportation system, and it is also a challenging task because of the underlying complex Spatio-temporal patterns and heterogeneities evolving across time. However, most present works mostly concentrate on solely capturing Spatial-temporal dependency or extracting implicit similarity graphs, but the hybrid-granularity evolution is ignored in their modeling process. In this paper, we proposed a novel data-driven end-to-end framework, named Spatio-Temporal Aware Hybrid Graph Network (STAHGNet), to couple the hybrid-grained heterogeneous correlations in series simultaneously through an elaborately Hybrid Graph Attention Module (HGAT) and Coarse-granularity Temporal Graph (CTG) generator. Furthermore, an automotive feature engineering with domain knowledge and a random neighbor sampling strategy is utilized to improve efficiency and reduce computational complexity. The MAE, RMSE, and MAPE are used for evaluation metrics. Tested on four real-life datasets, our proposal outperforms eight classical baselines and four state-of-the-art (SOTA) methods (e.g., MAE 14.82 on PeMSD3; MAE 18.92 on PeMSD4). Besides, extensive experiments and visualizations verify the effectiveness of each component in STAHGNet. In terms of computational cost, STAHGNet saves at least four times the space compared to the previous SOTA models. The proposed model will be beneficial for more efficient TFP as well as intelligent transport system construction.

Figures (8)

• Figure 1: The possible dynamic interactions of variables in TFP given different scales of observation. The red box means the sliding horizontal window, and vectors with different colors indicate representation vectors of different series learned by the model.
• Figure 2: The overall architecture of proposed STAHGNet. The single-step prediction is taken as the illustration example.
• Figure 3: Structure of STAHGNet cell. '$//$' on '$\rightarrow$' present stop-gradient.
• Figure 4: Structure of CTG generator.
• Figure 5: Overall Performance Comparison of Different Methods.