DeCaf: A Causal Decoupling Framework for OOD Generalization on Node Classification

Xiaoxue Han, Huzefa Rangwala, Yue Ning
Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, PMLR 258:2332-2340, 2025.

Abstract

Graph Neural Networks (GNNs) are susceptible to distribution shifts, creating vulnerability and security issues in critical domains. There is a pressing need to enhance the generalizability of GNNs on out-of-distribution (OOD) test data. Existing methods that target learning an invariant (feature, structure)-label mapping often depend on oversimplified assumptions about the data generation process, which do not adequately reflect the actual dynamics of distribution shifts in graphs. In this paper, we introduce a more realistic graph data generation model using Structural Causal Models (SCMs), allowing us to redefine distribution shifts by pinpointing their origins within the generation process. Building on this, we propose a casual decoupling framework, DeCaf, that independently learns unbiased feature-label and structure-label mappings. We provide a detailed theoretical framework that shows how our approach can effectively mitigate the impact of various distribution shifts. We evaluate DeCaf across both real-world and synthetic datasets that demonstrate different patterns of shifts, confirming its efficacy in enhancing the generalizability of GNNs. Our code is available at: \url{https://github.com/hanxiaoxue114/DeCaf-GraphOOD.}

Cite this Paper

BibTeX
@InProceedings{pmlr-v258-han25b, title = {DeCaf: A Causal Decoupling Framework for OOD Generalization on Node Classification}, author = {Han, Xiaoxue and Rangwala, Huzefa and Ning, Yue}, booktitle = {Proceedings of The 28th International Conference on Artificial Intelligence and Statistics}, pages = {2332--2340}, year = {2025}, editor = {Li, Yingzhen and Mandt, Stephan and Agrawal, Shipra and Khan, Emtiyaz}, volume = {258}, series = {Proceedings of Machine Learning Research}, month = {03--05 May}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v258/main/assets/han25b/han25b.pdf}, url = {https://proceedings.mlr.press/v258/han25b.html}, abstract = {Graph Neural Networks (GNNs) are susceptible to distribution shifts, creating vulnerability and security issues in critical domains. There is a pressing need to enhance the generalizability of GNNs on out-of-distribution (OOD) test data. Existing methods that target learning an invariant (feature, structure)-label mapping often depend on oversimplified assumptions about the data generation process, which do not adequately reflect the actual dynamics of distribution shifts in graphs. In this paper, we introduce a more realistic graph data generation model using Structural Causal Models (SCMs), allowing us to redefine distribution shifts by pinpointing their origins within the generation process. Building on this, we propose a casual decoupling framework, DeCaf, that independently learns unbiased feature-label and structure-label mappings. We provide a detailed theoretical framework that shows how our approach can effectively mitigate the impact of various distribution shifts. We evaluate DeCaf across both real-world and synthetic datasets that demonstrate different patterns of shifts, confirming its efficacy in enhancing the generalizability of GNNs. Our code is available at: \url{https://github.com/hanxiaoxue114/DeCaf-GraphOOD.}} }
Endnote
%0 Conference Paper %T DeCaf: A Causal Decoupling Framework for OOD Generalization on Node Classification %A Xiaoxue Han %A Huzefa Rangwala %A Yue Ning %B Proceedings of The 28th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2025 %E Yingzhen Li %E Stephan Mandt %E Shipra Agrawal %E Emtiyaz Khan %F pmlr-v258-han25b %I PMLR %P 2332--2340 %U https://proceedings.mlr.press/v258/han25b.html %V 258 %X Graph Neural Networks (GNNs) are susceptible to distribution shifts, creating vulnerability and security issues in critical domains. There is a pressing need to enhance the generalizability of GNNs on out-of-distribution (OOD) test data. Existing methods that target learning an invariant (feature, structure)-label mapping often depend on oversimplified assumptions about the data generation process, which do not adequately reflect the actual dynamics of distribution shifts in graphs. In this paper, we introduce a more realistic graph data generation model using Structural Causal Models (SCMs), allowing us to redefine distribution shifts by pinpointing their origins within the generation process. Building on this, we propose a casual decoupling framework, DeCaf, that independently learns unbiased feature-label and structure-label mappings. We provide a detailed theoretical framework that shows how our approach can effectively mitigate the impact of various distribution shifts. We evaluate DeCaf across both real-world and synthetic datasets that demonstrate different patterns of shifts, confirming its efficacy in enhancing the generalizability of GNNs. Our code is available at: \url{https://github.com/hanxiaoxue114/DeCaf-GraphOOD.}
APA
Han, X., Rangwala, H. & Ning, Y.. (2025). DeCaf: A Causal Decoupling Framework for OOD Generalization on Node Classification. Proceedings of The 28th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 258:2332-2340 Available from https://proceedings.mlr.press/v258/han25b.html.

Related Material