BeMap: Balanced Message Passing for Fair Graph Neural Network

Xiao Lin, Jian Kang, Weilin Cong, Hanghang Tong
Proceedings of the Second Learning on Graphs Conference, PMLR 231:37:1-37:25, 2024.

Abstract

Fairness in graph neural networks has been actively studied recently. However, existing works often do not explicitly consider the role of message passing in introducing or amplifying the bias. In this paper, we first investigate the problem of bias amplification in message passing. We empirically and theoretically demonstrate that message passing could amplify the bias when the 1-hop neighbors from different demographic groups are unbalanced. Guided by such analyses, we propose BeMap, a fair message passing method, that leverages a balance-aware sampling strategy to balance the number of the 1-hop neighbors of each node among different demographic groups. Extensive experiments on node classification demonstrate the efficacy of BeMap in mitigating bias while maintaining classification accuracy.

Cite this Paper

BibTeX
@InProceedings{pmlr-v231-lin24a, title = {BeMap: Balanced Message Passing for Fair Graph Neural Network}, author = {Lin, Xiao and Kang, Jian and Cong, Weilin and Tong, Hanghang}, booktitle = {Proceedings of the Second Learning on Graphs Conference}, pages = {37:1--37:25}, year = {2024}, editor = {Villar, Soledad and Chamberlain, Benjamin}, volume = {231}, series = {Proceedings of Machine Learning Research}, month = {27--30 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v231/lin24a/lin24a.pdf}, url = {https://proceedings.mlr.press/v231/lin24a.html}, abstract = {Fairness in graph neural networks has been actively studied recently. However, existing works often do not explicitly consider the role of message passing in introducing or amplifying the bias. In this paper, we first investigate the problem of bias amplification in message passing. We empirically and theoretically demonstrate that message passing could amplify the bias when the 1-hop neighbors from different demographic groups are unbalanced. Guided by such analyses, we propose BeMap, a fair message passing method, that leverages a balance-aware sampling strategy to balance the number of the 1-hop neighbors of each node among different demographic groups. Extensive experiments on node classification demonstrate the efficacy of BeMap in mitigating bias while maintaining classification accuracy.} }
Endnote
%0 Conference Paper %T BeMap: Balanced Message Passing for Fair Graph Neural Network %A Xiao Lin %A Jian Kang %A Weilin Cong %A Hanghang Tong %B Proceedings of the Second Learning on Graphs Conference %C Proceedings of Machine Learning Research %D 2024 %E Soledad Villar %E Benjamin Chamberlain %F pmlr-v231-lin24a %I PMLR %P 37:1--37:25 %U https://proceedings.mlr.press/v231/lin24a.html %V 231 %X Fairness in graph neural networks has been actively studied recently. However, existing works often do not explicitly consider the role of message passing in introducing or amplifying the bias. In this paper, we first investigate the problem of bias amplification in message passing. We empirically and theoretically demonstrate that message passing could amplify the bias when the 1-hop neighbors from different demographic groups are unbalanced. Guided by such analyses, we propose BeMap, a fair message passing method, that leverages a balance-aware sampling strategy to balance the number of the 1-hop neighbors of each node among different demographic groups. Extensive experiments on node classification demonstrate the efficacy of BeMap in mitigating bias while maintaining classification accuracy.
APA
Lin, X., Kang, J., Cong, W. & Tong, H.. (2024). BeMap: Balanced Message Passing for Fair Graph Neural Network. Proceedings of the Second Learning on Graphs Conference, in Proceedings of Machine Learning Research 231:37:1-37:25 Available from https://proceedings.mlr.press/v231/lin24a.html.

Related Material