Feature Importance Disparities for Data Bias Investigations

Peter W Chang, Leor Fishman, Seth Neel
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:6170-6201, 2024.

Abstract

It is widely held that one cause of downstream bias in classifiers is bias present in the training data. Rectifying such biases may involve context-dependent interventions such as training separate models on subgroups, removing features with bias in the collection process, or even conducting real-world experiments to ascertain sources of bias. Despite the need for such data bias investigations, few automated methods exist to assist practitioners in these efforts. In this paper, we present one such method that given a dataset $X$ consisting of protected and unprotected features, outcomes $y$, and a regressor $h$ that predicts $y$ given $X$, outputs a tuple $(f_j, g)$, with the following property: $g$ corresponds to a subset of the training dataset $(X, y)$, such that the $j^{th}$ feature $f_j$ has much larger (or smaller) influence in the subgroup $g$, than on the dataset overall, which we call feature importance disparity (FID). We show across $4$ datasets and $4$ common feature importance methods of broad interest to the machine learning community that we can efficiently find subgroups with large FID values even over exponentially large subgroup classes and in practice these groups correspond to subgroups with potentially serious bias issues as measured by standard fairness metrics.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-chang24a, title = {Feature Importance Disparities for Data Bias Investigations}, author = {Chang, Peter W and Fishman, Leor and Neel, Seth}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {6170--6201}, 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/chang24a/chang24a.pdf}, url = {https://proceedings.mlr.press/v235/chang24a.html}, abstract = {It is widely held that one cause of downstream bias in classifiers is bias present in the training data. Rectifying such biases may involve context-dependent interventions such as training separate models on subgroups, removing features with bias in the collection process, or even conducting real-world experiments to ascertain sources of bias. Despite the need for such data bias investigations, few automated methods exist to assist practitioners in these efforts. In this paper, we present one such method that given a dataset $X$ consisting of protected and unprotected features, outcomes $y$, and a regressor $h$ that predicts $y$ given $X$, outputs a tuple $(f_j, g)$, with the following property: $g$ corresponds to a subset of the training dataset $(X, y)$, such that the $j^{th}$ feature $f_j$ has much larger (or smaller) influence in the subgroup $g$, than on the dataset overall, which we call feature importance disparity (FID). We show across $4$ datasets and $4$ common feature importance methods of broad interest to the machine learning community that we can efficiently find subgroups with large FID values even over exponentially large subgroup classes and in practice these groups correspond to subgroups with potentially serious bias issues as measured by standard fairness metrics.} }
Endnote
%0 Conference Paper %T Feature Importance Disparities for Data Bias Investigations %A Peter W Chang %A Leor Fishman %A Seth Neel %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-chang24a %I PMLR %P 6170--6201 %U https://proceedings.mlr.press/v235/chang24a.html %V 235 %X It is widely held that one cause of downstream bias in classifiers is bias present in the training data. Rectifying such biases may involve context-dependent interventions such as training separate models on subgroups, removing features with bias in the collection process, or even conducting real-world experiments to ascertain sources of bias. Despite the need for such data bias investigations, few automated methods exist to assist practitioners in these efforts. In this paper, we present one such method that given a dataset $X$ consisting of protected and unprotected features, outcomes $y$, and a regressor $h$ that predicts $y$ given $X$, outputs a tuple $(f_j, g)$, with the following property: $g$ corresponds to a subset of the training dataset $(X, y)$, such that the $j^{th}$ feature $f_j$ has much larger (or smaller) influence in the subgroup $g$, than on the dataset overall, which we call feature importance disparity (FID). We show across $4$ datasets and $4$ common feature importance methods of broad interest to the machine learning community that we can efficiently find subgroups with large FID values even over exponentially large subgroup classes and in practice these groups correspond to subgroups with potentially serious bias issues as measured by standard fairness metrics.
APA
Chang, P.W., Fishman, L. & Neel, S.. (2024). Feature Importance Disparities for Data Bias Investigations. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:6170-6201 Available from https://proceedings.mlr.press/v235/chang24a.html.

Related Material