cs.LGJul 14, 2024

Improving Graph Out-of-distribution Generalization Beyond Causality

Authors: Can Xu, Yao Cheng, Jianxiang Yu, Haosen Wang, Jingsong Lv, Yao Liu, Xiang Li

Organizations: East China Normal University Shanghai, China · Southeast University Nanjing, China · Zhejiang Lab Hangzhou, China

Abstract

Existing methods for graph out-of-distribution (OOD) generalization primarily rely on empirical studies on synthetic datasets. Such approaches tend to overemphasize the causal relationships between invariant sub-graphs and labels, thereby neglecting the non-negligible role of environment in real-world scenarios. In contrast to previous studies that impose rigid independence assumptions on environments and invariant sub-graphs, this paper presents the theorems of environment-label dependency and mutable rationale invariance, where the former characterizes the usefulness of environments in determining graph labels while the latter refers to the mutable importance of graph rationales. Based on analytic investigations, a novel variational inference based method named ``Probability Dependency on Environments and Rationales for OOD Graphs on Real-world Data'' (DEROG) is introduced. To alleviate the adverse effect of unknown prior knowledge on environments and rationales, DEROG utilizes generalized Bayesian inference. Further, DEROG employs an EM-based algorithm for optimization. Finally, extensive experiments on real-world datasets under different distribution shifts are conducted to show the superiority of DEROG. Our code is publicly available at https://github.com/LEOXC1571/DEROG.

Figures & tables

Explore similar work

CardsList
  1. Adversarial Label Invariant Graph Data Augmentations for Out-of-Distribution Generalization

    Apr 9, 2026Simon Zhang, Ryan P. DeMilt, Kun Jin +1OOD GeneralizationCovariate Shift

  2. Implicit Feature Reweighting for Label-Free Graph Generalization

    Oct 21, 2026Tianjun Yao, Chen Xi, Xinyang Li +1OOD GeneralizationGraph Representation Learning

  3. Towards OOD Generalization in Dynamic Graphs via Causal Invariant Learning

    Mar 2, 2026Xinxun Zhang, Pengfei Jiao, Mengzhou Gao +2OOD GeneralizationDynamic Graph Learning