EUVIMEDEUVIMED entdecken
Crossref

Towards OOD Generalization in Dynamic Graphs via Causal Invariant Learning

Xinxun Zhang, Pengfei Jiao, Mengzhou Gao, Tianpeng Li, Xuan Guo

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (19) · 16379-16387

DOI: 10.1609/aaai.v40i19.38676 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Xinxun Zhang und 4 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Although dynamic graph neural networks (DyGNNs) have demonstrated promising capabilities, most existing methods ignore out-of-distribution (OOD) shifts that commonly exist in dynamic graphs. Dynamic graph OOD generalization is non-trivial due to the following challenges: 1) Identifying invariant and variant patterns amid complex graph evolution, 2) Capturing the intrinsic evolution rationale from these patterns, and 3) Ensuring model generalization across diverse OOD shifts despite limited data distribution observations. Although several attempts have been made to tackle these challenges, none has successfully addressed all three simultaneously, and they face various limitations in complex OOD scenarios. To solve these issues, we propose a Dynamic graph Causal Invariant Learning (DyCIL) model for OOD generalization via exploiting invariant spatio-temporal patterns from a causal view. Specifically, we first develop a dynamic causal subgraph generator to identify causal dynamic subgraphs explicitly. Next, we design a causal-aware spatio-temporal attention module to extract the intrinsic evolution rationale behind invariant patterns. Finally, we further introduce an adaptive environment generator to capture the underlying dynamics of distributional shifts. Extensive experiments on both real-world and synthetic dynamic graph datasets demonstrate the superiority of our model over state-of-the-art baselines in handling OOD shifts.

Zitieren

APA

Zhang, X., Jiao, P., Gao, M., Li, T., & Guo, X. (2026). Towards OOD Generalization in Dynamic Graphs via Causal Invariant Learning. Proceedings of the AAAI Conference on Artificial Intelligence, 40(19), 16379-16387. https://doi.org/10.1609/aaai.v40i19.38676

Vancouver

Zhang X, Jiao P, Gao M, Li T, Guo X. Towards OOD Generalization in Dynamic Graphs via Causal Invariant Learning. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(19):16379-16387. doi:10.1609/aaai.v40i19.38676

BibTeX
@article{Zhang2026,
  author = {Zhang, Xinxun and Jiao, Pengfei and Gao, Mengzhou and Li, Tianpeng and Guo, Xuan},
  title = {Towards OOD Generalization in Dynamic Graphs via Causal Invariant Learning},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {19},
  pages = {16379--16387},
  doi = {10.1609/aaai.v40i19.38676}
}
Kostenlos starten →