Simple and near-optimal algorithms for hidden stratification and multi-group learning

Christopher J Tosh, Daniel Hsu
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:21633-21657, 2022.

Abstract

Multi-group agnostic learning is a formal learning criterion that is concerned with the conditional risks of predictors within subgroups of a population. The criterion addresses recent practical concerns such as subgroup fairness and hidden stratification. This paper studies the structure of solutions to the multi-group learning problem, and provides simple and near-optimal algorithms for the learning problem.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-tosh22a, title = {Simple and near-optimal algorithms for hidden stratification and multi-group learning}, author = {Tosh, Christopher J and Hsu, Daniel}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {21633--21657}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/tosh22a/tosh22a.pdf}, url = {https://proceedings.mlr.press/v162/tosh22a.html}, abstract = {Multi-group agnostic learning is a formal learning criterion that is concerned with the conditional risks of predictors within subgroups of a population. The criterion addresses recent practical concerns such as subgroup fairness and hidden stratification. This paper studies the structure of solutions to the multi-group learning problem, and provides simple and near-optimal algorithms for the learning problem.} }
Endnote
%0 Conference Paper %T Simple and near-optimal algorithms for hidden stratification and multi-group learning %A Christopher J Tosh %A Daniel Hsu %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-tosh22a %I PMLR %P 21633--21657 %U https://proceedings.mlr.press/v162/tosh22a.html %V 162 %X Multi-group agnostic learning is a formal learning criterion that is concerned with the conditional risks of predictors within subgroups of a population. The criterion addresses recent practical concerns such as subgroup fairness and hidden stratification. This paper studies the structure of solutions to the multi-group learning problem, and provides simple and near-optimal algorithms for the learning problem.
APA
Tosh, C.J. & Hsu, D.. (2022). Simple and near-optimal algorithms for hidden stratification and multi-group learning. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:21633-21657 Available from https://proceedings.mlr.press/v162/tosh22a.html.

Related Material