CELL: A Causal Perspective for Fairness-aware Graph Adaptation

Hourun Li, Yifan Wang, Qinghua Ran, Junyu Luo, Jia Yang, Changling Zhou, Zhiping Xiao, Wei Ju, Xiao Luo, Ming Zhang
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:70878-70896, 2026.

Abstract

This paper studies fairness-aware graph adaptation, aiming to transfer knowledge from a labeled source graph to an unlabeled target graph while addressing fairness. Most prior methods require target-domain attributes to learn invariant graph representations of sensitive attributes, which are often unavailable in practice. To address this limitation, we introduce Causality-attended Representation Dientanglement with Structural Alignment (CELL) for fairness-aware graph adaptation without requiring target sensitive labels. CELL constructs a causal graph to model the graph-generation mechanism and guide fair representation disentanglement. Specifically, CELL uses sensitive and causal encoders to extract sensitive and causal factors, respectively, and promotes disentanglement by minimizing their conditional mutual information. To leverage unlabeled target data, we further generate pseudo-labels for both target task labels and sensitive attributes, and use similarity relations to derive unbiased node representations. Finally, to further mitigate domain shift, we build a fairness-aware bipartite graph that provides additional structural supervision for cross-domain alignment. Experiments on benchmarks show that CELL consistently outperforms strong baselines in both predictive performance and fairness.

Cite this Paper


BibTeX
@InProceedings{pmlr-v306-li26gu, title = {{CELL}: A Causal Perspective for Fairness-aware Graph Adaptation}, author = {Li, Hourun and Wang, Yifan and Ran, Qinghua and Luo, Junyu and Yang, Jia and Zhou, Changling and Xiao, Zhiping and Ju, Wei and Luo, Xiao and Zhang, Ming}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {70878--70896}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/li26gu/li26gu.pdf}, url = {https://proceedings.mlr.press/v306/li26gu.html}, abstract = {This paper studies fairness-aware graph adaptation, aiming to transfer knowledge from a labeled source graph to an unlabeled target graph while addressing fairness. Most prior methods require target-domain attributes to learn invariant graph representations of sensitive attributes, which are often unavailable in practice. To address this limitation, we introduce Causality-attended Representation Dientanglement with Structural Alignment (CELL) for fairness-aware graph adaptation without requiring target sensitive labels. CELL constructs a causal graph to model the graph-generation mechanism and guide fair representation disentanglement. Specifically, CELL uses sensitive and causal encoders to extract sensitive and causal factors, respectively, and promotes disentanglement by minimizing their conditional mutual information. To leverage unlabeled target data, we further generate pseudo-labels for both target task labels and sensitive attributes, and use similarity relations to derive unbiased node representations. Finally, to further mitigate domain shift, we build a fairness-aware bipartite graph that provides additional structural supervision for cross-domain alignment. Experiments on benchmarks show that CELL consistently outperforms strong baselines in both predictive performance and fairness.} }
Endnote
%0 Conference Paper %T CELL: A Causal Perspective for Fairness-aware Graph Adaptation %A Hourun Li %A Yifan Wang %A Qinghua Ran %A Junyu Luo %A Jia Yang %A Changling Zhou %A Zhiping Xiao %A Wei Ju %A Xiao Luo %A Ming Zhang %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-li26gu %I PMLR %P 70878--70896 %U https://proceedings.mlr.press/v306/li26gu.html %V 306 %X This paper studies fairness-aware graph adaptation, aiming to transfer knowledge from a labeled source graph to an unlabeled target graph while addressing fairness. Most prior methods require target-domain attributes to learn invariant graph representations of sensitive attributes, which are often unavailable in practice. To address this limitation, we introduce Causality-attended Representation Dientanglement with Structural Alignment (CELL) for fairness-aware graph adaptation without requiring target sensitive labels. CELL constructs a causal graph to model the graph-generation mechanism and guide fair representation disentanglement. Specifically, CELL uses sensitive and causal encoders to extract sensitive and causal factors, respectively, and promotes disentanglement by minimizing their conditional mutual information. To leverage unlabeled target data, we further generate pseudo-labels for both target task labels and sensitive attributes, and use similarity relations to derive unbiased node representations. Finally, to further mitigate domain shift, we build a fairness-aware bipartite graph that provides additional structural supervision for cross-domain alignment. Experiments on benchmarks show that CELL consistently outperforms strong baselines in both predictive performance and fairness.
APA
Li, H., Wang, Y., Ran, Q., Luo, J., Yang, J., Zhou, C., Xiao, Z., Ju, W., Luo, X. & Zhang, M.. (2026). CELL: A Causal Perspective for Fairness-aware Graph Adaptation. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:70878-70896 Available from https://proceedings.mlr.press/v306/li26gu.html.

Related Material