Time-Space-Interlaced Spatiotemporal Graph Forecasting via Two-Stage Summarized Attention
Lists
Only you see this
Abstract
Typical spatiotemporal graph forecasting methods process graph-structured spatiotemporal data respectively from spatial and temporal perspectives with the idea of divide and conquer. Existing works are incapable of capturing long-term transdimensional correlations among different spatial points in different time planes, i.e., time-space-interlaced correlations. To tackle this issue, we propose a two-stage summarized attention network to establish transdimensional direct message passing routes between different data points in different time planes and spaces, thus enabling the extraction of time-space-interlaced long-term correlations. Specifically, a novel spatiotemporal embedding is proposed to implement time-space-interlaced learning by expanding orthogonal spatial and temporal dimensionalities into one-dimensionality, a series of temporal context fusion units are added to address the fluctuation dislocation insensitivity issue which is caused by time-space-dimension expansion, and an ingenious two-stage design can significantly reduce the computation complexity of such time-space-interlaced learning. Extensive experiments illustrate the superior performance of our proposed approach on real-world spatiotemporal datasets.
BibTeX
@inproceedings{sun2025time,
author = {Zhaoyang Sun and Yudong Zhang and Xuan Yu and Kai Wang and Binwu Wang and Yang Wang and Xu Wang},
title = {{Time-Space-Interlaced Spatiotemporal Graph Forecasting via Two-Stage Summarized Attention}},
booktitle = {IEEE International Conference on Acoustics, Speech and Signal Processing},
year = {2025},
doi = {10.1109/icassp49660.2025.10889478},
}Loading related content...