Meta Optimality for Demographic Parity Constrained Regression via Post-Processing

Kazuto Fukuchi
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:18024-18046, 2025.

Abstract

We address the regression problem under the constraint of demographic parity, a commonly used fairness definition. Recent studies have revealed fair minimax optimal regression algorithms, the most accurate algorithms that adhere to the fairness constraint. However, these analyses are tightly coupled with specific data generation models. In this paper, we provide meta-theorems that can be applied to various situations to validate the fair minimax optimality of the corresponding regression algorithms. Furthermore, we demonstrate that fair minimax optimal regression can be achieved through post-processing methods, allowing researchers and practitioners to focus on improving conventional regression techniques, which can then be efficiently adapted for fair regression.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-fukuchi25a, title = {Meta Optimality for Demographic Parity Constrained Regression via Post-Processing}, author = {Fukuchi, Kazuto}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {18024--18046}, 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/fukuchi25a/fukuchi25a.pdf}, url = {https://proceedings.mlr.press/v267/fukuchi25a.html}, abstract = {We address the regression problem under the constraint of demographic parity, a commonly used fairness definition. Recent studies have revealed fair minimax optimal regression algorithms, the most accurate algorithms that adhere to the fairness constraint. However, these analyses are tightly coupled with specific data generation models. In this paper, we provide meta-theorems that can be applied to various situations to validate the fair minimax optimality of the corresponding regression algorithms. Furthermore, we demonstrate that fair minimax optimal regression can be achieved through post-processing methods, allowing researchers and practitioners to focus on improving conventional regression techniques, which can then be efficiently adapted for fair regression.} }
Endnote
%0 Conference Paper %T Meta Optimality for Demographic Parity Constrained Regression via Post-Processing %A Kazuto Fukuchi %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-fukuchi25a %I PMLR %P 18024--18046 %U https://proceedings.mlr.press/v267/fukuchi25a.html %V 267 %X We address the regression problem under the constraint of demographic parity, a commonly used fairness definition. Recent studies have revealed fair minimax optimal regression algorithms, the most accurate algorithms that adhere to the fairness constraint. However, these analyses are tightly coupled with specific data generation models. In this paper, we provide meta-theorems that can be applied to various situations to validate the fair minimax optimality of the corresponding regression algorithms. Furthermore, we demonstrate that fair minimax optimal regression can be achieved through post-processing methods, allowing researchers and practitioners to focus on improving conventional regression techniques, which can then be efficiently adapted for fair regression.
APA
Fukuchi, K.. (2025). Meta Optimality for Demographic Parity Constrained Regression via Post-Processing. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:18024-18046 Available from https://proceedings.mlr.press/v267/fukuchi25a.html.

Related Material