Intersectional Unfairness Discovery

Gezheng Xu, Qi Chen, Charles Ling, Boyu Wang, Changjian Shui
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:54888-54917, 2024.

Abstract

AI systems have been shown to produce unfair results for certain subgroups of population, highlighting the need to understand bias on certain sensitive attributes. Current research often falls short, primarily focusing on the subgroups characterized by a single sensitive attribute, while neglecting the nature of intersectional fairness of multiple sensitive attributes. This paper focuses on its one fundamental aspect by discovering diverse high-bias intersectional sensitive attributes. Specifically, we propose a Bias-Guided Generative Network (BGGN). By treating each bias value as a reward, BGGN efficiently generates high-bias intersectional sensitive attributes. Experiments on real-world text and image datasets demonstrate a diverse and efficient discovery of BGGN. To further evaluate the generated unseen but possible unfair intersectional sensitive attributes, we formulate them as prompts and use modern generative AI to produce new text and images. The results of frequently generating biased data provides new insights of discovering potential unfairness in popular modern generative AI systems. Warning: This paper contains examples that are offensive in nature.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-xu24d, title = {Intersectional Unfairness Discovery}, author = {Xu, Gezheng and Chen, Qi and Ling, Charles and Wang, Boyu and Shui, Changjian}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {54888--54917}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/xu24d/xu24d.pdf}, url = {https://proceedings.mlr.press/v235/xu24d.html}, abstract = {AI systems have been shown to produce unfair results for certain subgroups of population, highlighting the need to understand bias on certain sensitive attributes. Current research often falls short, primarily focusing on the subgroups characterized by a single sensitive attribute, while neglecting the nature of intersectional fairness of multiple sensitive attributes. This paper focuses on its one fundamental aspect by discovering diverse high-bias intersectional sensitive attributes. Specifically, we propose a Bias-Guided Generative Network (BGGN). By treating each bias value as a reward, BGGN efficiently generates high-bias intersectional sensitive attributes. Experiments on real-world text and image datasets demonstrate a diverse and efficient discovery of BGGN. To further evaluate the generated unseen but possible unfair intersectional sensitive attributes, we formulate them as prompts and use modern generative AI to produce new text and images. The results of frequently generating biased data provides new insights of discovering potential unfairness in popular modern generative AI systems. Warning: This paper contains examples that are offensive in nature.} }
Endnote
%0 Conference Paper %T Intersectional Unfairness Discovery %A Gezheng Xu %A Qi Chen %A Charles Ling %A Boyu Wang %A Changjian Shui %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-xu24d %I PMLR %P 54888--54917 %U https://proceedings.mlr.press/v235/xu24d.html %V 235 %X AI systems have been shown to produce unfair results for certain subgroups of population, highlighting the need to understand bias on certain sensitive attributes. Current research often falls short, primarily focusing on the subgroups characterized by a single sensitive attribute, while neglecting the nature of intersectional fairness of multiple sensitive attributes. This paper focuses on its one fundamental aspect by discovering diverse high-bias intersectional sensitive attributes. Specifically, we propose a Bias-Guided Generative Network (BGGN). By treating each bias value as a reward, BGGN efficiently generates high-bias intersectional sensitive attributes. Experiments on real-world text and image datasets demonstrate a diverse and efficient discovery of BGGN. To further evaluate the generated unseen but possible unfair intersectional sensitive attributes, we formulate them as prompts and use modern generative AI to produce new text and images. The results of frequently generating biased data provides new insights of discovering potential unfairness in popular modern generative AI systems. Warning: This paper contains examples that are offensive in nature.
APA
Xu, G., Chen, Q., Ling, C., Wang, B. & Shui, C.. (2024). Intersectional Unfairness Discovery. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:54888-54917 Available from https://proceedings.mlr.press/v235/xu24d.html.

Related Material