Robust Spatio-Temporal Centralized Interaction for OOD Learning

Jiaming Ma, Binwu Wang, Pengkun Wang, Zhengyang Zhou, Xu Wang, Yang Wang
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:42165-42192, 2025.

Abstract

Recently, spatiotemporal graph convolutional networks have achieved dominant performance in spatiotemporal prediction tasks. However, most models relying on node-to-node messaging interaction exhibit sensitivity to spatiotemporal shifts, encountering out-of-distribution (OOD) challenges. To address these issues, we introduce Spatio-Temporal OOD Processor (STOP), which employs a centralized messaging mechanism along with a message perturbation mechanism to facilitate robust spatiotemporal interactions. Specifically, the centralized messaging mechanism integrates Context-Aware Units for coarse-grained spatiotemporal feature interactions with nodes, effectively blocking traditional node-to-node messages. We also implement a message perturbation mechanism to disrupt this messaging process, compelling the model to extract generalizable contextual features from generated variant environments. Finally, we customize a spatiotemporal distributionally robust optimization approach that exposes the model to challenging environments, thereby further enhancing its generalization capabilities. Compared with 14 baselines across six datasets, STOP achieves up to 17.01% improvement in generalization performance and 18.44% improvement in inductive learning performance. The code is available at https://github.com/PoorOtterBob/STOP.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-ma25s, title = {Robust Spatio-Temporal Centralized Interaction for {OOD} Learning}, author = {Ma, Jiaming and Wang, Binwu and Wang, Pengkun and Zhou, Zhengyang and Wang, Xu and Wang, Yang}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {42165--42192}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/ma25s/ma25s.pdf}, url = {https://proceedings.mlr.press/v267/ma25s.html}, abstract = {Recently, spatiotemporal graph convolutional networks have achieved dominant performance in spatiotemporal prediction tasks. However, most models relying on node-to-node messaging interaction exhibit sensitivity to spatiotemporal shifts, encountering out-of-distribution (OOD) challenges. To address these issues, we introduce Spatio-Temporal OOD Processor (STOP), which employs a centralized messaging mechanism along with a message perturbation mechanism to facilitate robust spatiotemporal interactions. Specifically, the centralized messaging mechanism integrates Context-Aware Units for coarse-grained spatiotemporal feature interactions with nodes, effectively blocking traditional node-to-node messages. We also implement a message perturbation mechanism to disrupt this messaging process, compelling the model to extract generalizable contextual features from generated variant environments. Finally, we customize a spatiotemporal distributionally robust optimization approach that exposes the model to challenging environments, thereby further enhancing its generalization capabilities. Compared with 14 baselines across six datasets, STOP achieves up to 17.01% improvement in generalization performance and 18.44% improvement in inductive learning performance. The code is available at https://github.com/PoorOtterBob/STOP.} }
Endnote
%0 Conference Paper %T Robust Spatio-Temporal Centralized Interaction for OOD Learning %A Jiaming Ma %A Binwu Wang %A Pengkun Wang %A Zhengyang Zhou %A Xu Wang %A Yang Wang %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-ma25s %I PMLR %P 42165--42192 %U https://proceedings.mlr.press/v267/ma25s.html %V 267 %X Recently, spatiotemporal graph convolutional networks have achieved dominant performance in spatiotemporal prediction tasks. However, most models relying on node-to-node messaging interaction exhibit sensitivity to spatiotemporal shifts, encountering out-of-distribution (OOD) challenges. To address these issues, we introduce Spatio-Temporal OOD Processor (STOP), which employs a centralized messaging mechanism along with a message perturbation mechanism to facilitate robust spatiotemporal interactions. Specifically, the centralized messaging mechanism integrates Context-Aware Units for coarse-grained spatiotemporal feature interactions with nodes, effectively blocking traditional node-to-node messages. We also implement a message perturbation mechanism to disrupt this messaging process, compelling the model to extract generalizable contextual features from generated variant environments. Finally, we customize a spatiotemporal distributionally robust optimization approach that exposes the model to challenging environments, thereby further enhancing its generalization capabilities. Compared with 14 baselines across six datasets, STOP achieves up to 17.01% improvement in generalization performance and 18.44% improvement in inductive learning performance. The code is available at https://github.com/PoorOtterBob/STOP.
APA
Ma, J., Wang, B., Wang, P., Zhou, Z., Wang, X. & Wang, Y.. (2025). Robust Spatio-Temporal Centralized Interaction for OOD Learning. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:42165-42192 Available from https://proceedings.mlr.press/v267/ma25s.html.

Related Material