Spatio-temporal forecasting, a prominent research focus for understanding the dynamics of data flow in various domains, has recently extended its significance to traffic prediction as a notable application. Considering topology of data flow, existing methods mainly utilize Graph Convolutional Networks (GCNs), where graph construction is the basic concern and generally determines how data features are propagated. However, through extensive investigations and theoretical evidence, it is revealed that graph construction strategy, which is usually regarded as the key step to success, has actually provided very little benefit, while the presence of feature propagation itself on spatiotemporal domain is more significant, i.e., feature propagation carry more weights than how they do. Thus, by making slight refinements of a feature normalization method, we propose a Spatio-Temporal Propagation (STP) module, which does not require intervention of a specific graph structure, yet simple and effective. Various experiments on public datasets verify that the proposed STP module is an on-the-shelf tool that can be accessed to the end of current models or even replace GCNs as an alternative for capturing spatiotemporal features, while achieving better predictions. All the source codes are open accessed on GitHub.
Revisiting Spatio-Temporal Forecasting: Feature Propagation Carry More Weights Than How They Do
2024-09-24
717937 byte
Conference paper
Electronic Resource
English
MLP for Spatio-Temporal Traffic Volume Forecasting
IEEE | 2021
|STGT: FORECASTING PEDESTRIAN MOTION USING SPATIO-TEMPORAL GRAPH TRANSFORMER
British Library Conference Proceedings | 2021
|