ST-UNet: A Spatio-Temporal U-Network for Graph-structured Time Series Modeling

TL;DR

This paper introduces ST-UNet, a novel multi-scale U-shaped neural network designed for modeling complex, dynamic spatio-temporal graph data, effectively capturing multi-scale features for improved prediction accuracy.

Contribution

The paper presents a new multi-scale U-Net architecture with paired sampling operations for spatio-temporal graphs, enabling simultaneous spatial and temporal feature extraction.

Findings

Outperforms existing methods on real-world datasets

Effectively captures multi-scale spatio-temporal features

Achieves significant improvements in prediction accuracy

Abstract

The spatio-temporal graph learning is becoming an increasingly important object of graph study. Many application domains involve highly dynamic graphs where temporal information is crucial, e.g. traffic networks and financial transaction graphs. Despite the constant progress made on learning structured data, there is still a lack of effective means to extract dynamic complex features from spatio-temporal structures. Particularly, conventional models such as convolutional networks or recurrent neural networks are incapable of revealing the temporal patterns in short or long terms and exploring the spatial properties in local or global scope from spatio-temporal graphs simultaneously. To tackle this problem, we design a novel multi-scale architecture, Spatio-Temporal U-Net (ST-UNet), for graph-structured time series modeling. In this U-shaped network, a paired sampling operation is proposed in spacetime domain accordingly: the pooling (ST-Pool) coarsens the input graph in spatial from its deterministic partition while abstracts multi-resolution temporal dependencies through dilated recurrent skip connections; based on previous settings in the downsampling, the unpooling (ST-Unpool) restores the original structure of spatio-temporal graphs and resumes regular intervals within graph sequences. Experiments on spatio-temporal prediction tasks demonstrate that our model effectively captures comprehensive features in multiple scales and achieves substantial improvements over mainstream methods on several real-world datasets.