DANCE: Dual Unbiased Expansion with Group-acquired Alignment for Out-of-distribution Graph Fairness Learning

Yifan Wang, Hourun Li, Ling Yue, Zhiping Xiao, Jia Yang, Changling Zhou, Wei Ju, Ming Zhang, Xiao Luo
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:63516-63529, 2025.

Abstract

Graph neural networks (GNNs) have shown strong performance in graph fairness learning, which aims to ensure that predictions are unbiased with respect to sensitive attributes. However, existing approaches usually assume that training and test data share the same distribution, which rarely holds in the real world. To tackle this challenge, we propose a novel approach named Dual Unbiased Expansion with Group-acquired Alignment (DANCE) for graph fairness learning under distribution shifts. The core idea of our DANCE is to synthesize challenging yet unbiased virtual graph data in both graph and hidden spaces, simulating distribution shifts from a data-centric view. Specifically, we introduce the unbiased Mixup in the hidden space, prioritizing minor groups to address the potential imbalance of sensitive attributes. Simultaneously, we conduct fairness-aware adversarial learning in the graph space to focus on challenging samples and improve model robustness. To further bridge the domain gap, we propose a group-acquired alignment objective that prioritizes negative pair groups with identical sensitive labels. Additionally, a representation disentanglement objective is adopted to decorrelate sensitive attributes and target representations for enhanced fairness. Extensive experiments demonstrate the superior effectiveness of the proposed DANCE.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-wang25bf, title = {{DANCE}: Dual Unbiased Expansion with Group-acquired Alignment for Out-of-distribution Graph Fairness Learning}, author = {Wang, Yifan and Li, Hourun and Yue, Ling and Xiao, Zhiping and Yang, Jia and Zhou, Changling and Ju, Wei and Zhang, Ming and Luo, Xiao}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {63516--63529}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/wang25bf/wang25bf.pdf}, url = {https://proceedings.mlr.press/v267/wang25bf.html}, abstract = {Graph neural networks (GNNs) have shown strong performance in graph fairness learning, which aims to ensure that predictions are unbiased with respect to sensitive attributes. However, existing approaches usually assume that training and test data share the same distribution, which rarely holds in the real world. To tackle this challenge, we propose a novel approach named Dual Unbiased Expansion with Group-acquired Alignment (DANCE) for graph fairness learning under distribution shifts. The core idea of our DANCE is to synthesize challenging yet unbiased virtual graph data in both graph and hidden spaces, simulating distribution shifts from a data-centric view. Specifically, we introduce the unbiased Mixup in the hidden space, prioritizing minor groups to address the potential imbalance of sensitive attributes. Simultaneously, we conduct fairness-aware adversarial learning in the graph space to focus on challenging samples and improve model robustness. To further bridge the domain gap, we propose a group-acquired alignment objective that prioritizes negative pair groups with identical sensitive labels. Additionally, a representation disentanglement objective is adopted to decorrelate sensitive attributes and target representations for enhanced fairness. Extensive experiments demonstrate the superior effectiveness of the proposed DANCE.} }
Endnote
%0 Conference Paper %T DANCE: Dual Unbiased Expansion with Group-acquired Alignment for Out-of-distribution Graph Fairness Learning %A Yifan Wang %A Hourun Li %A Ling Yue %A Zhiping Xiao %A Jia Yang %A Changling Zhou %A Wei Ju %A Ming Zhang %A Xiao Luo %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-wang25bf %I PMLR %P 63516--63529 %U https://proceedings.mlr.press/v267/wang25bf.html %V 267 %X Graph neural networks (GNNs) have shown strong performance in graph fairness learning, which aims to ensure that predictions are unbiased with respect to sensitive attributes. However, existing approaches usually assume that training and test data share the same distribution, which rarely holds in the real world. To tackle this challenge, we propose a novel approach named Dual Unbiased Expansion with Group-acquired Alignment (DANCE) for graph fairness learning under distribution shifts. The core idea of our DANCE is to synthesize challenging yet unbiased virtual graph data in both graph and hidden spaces, simulating distribution shifts from a data-centric view. Specifically, we introduce the unbiased Mixup in the hidden space, prioritizing minor groups to address the potential imbalance of sensitive attributes. Simultaneously, we conduct fairness-aware adversarial learning in the graph space to focus on challenging samples and improve model robustness. To further bridge the domain gap, we propose a group-acquired alignment objective that prioritizes negative pair groups with identical sensitive labels. Additionally, a representation disentanglement objective is adopted to decorrelate sensitive attributes and target representations for enhanced fairness. Extensive experiments demonstrate the superior effectiveness of the proposed DANCE.
APA
Wang, Y., Li, H., Yue, L., Xiao, Z., Yang, J., Zhou, C., Ju, W., Zhang, M. & Luo, X.. (2025). DANCE: Dual Unbiased Expansion with Group-acquired Alignment for Out-of-distribution Graph Fairness Learning. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:63516-63529 Available from https://proceedings.mlr.press/v267/wang25bf.html.

Related Material