Robust Spatio-Temporal Centralized Interaction for OOD Learning

Recently, spatio-temporal graph convolutional networks have achieved dominant performance in spatio-temporal prediction tasks. However, most models relying on node-to-node messaging interaction exhibit sensitivity to spatio-temporal 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 spatio-temporal interactions. Specifically, the centralized messaging mechanism integrates Context-Aware Units for coarse-grained spatio-temporal 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 spatio-temporal 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 .