Learning Non-Discriminatory Predictors

Blake Woodworth, Suriya Gunasekar, Mesrob I. Ohannessian, Nathan Srebro
Proceedings of the 2017 Conference on Learning Theory, PMLR 65:1920-1953, 2017.

Abstract

We consider learning a predictor which is non-discriminatory with respect to a “protected attribute” according to the notion of “equalized odds” proposed by Hardt et al. (2016). We study the problem of learning such a non-discriminatory predictor from a finite training set, both statistically and computationally. We show that a post-hoc correction approach, as suggested by Hardt et al, can be highly suboptimal, present a nearly-optimal statistical procedure, argue that the associated computational problem is intractable, and suggest a second moment relaxation of the non-discrimination definition for which learning is tractable.

Cite this Paper

BibTeX
@InProceedings{pmlr-v65-woodworth17a, title = {Learning Non-Discriminatory Predictors}, author = {Woodworth, Blake and Gunasekar, Suriya and Ohannessian, Mesrob I. and Srebro, Nathan}, booktitle = {Proceedings of the 2017 Conference on Learning Theory}, pages = {1920--1953}, year = {2017}, editor = {Kale, Satyen and Shamir, Ohad}, volume = {65}, series = {Proceedings of Machine Learning Research}, month = {07--10 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v65/woodworth17a/woodworth17a.pdf}, url = {https://proceedings.mlr.press/v65/woodworth17a.html}, abstract = {We consider learning a predictor which is non-discriminatory with respect to a “protected attribute” according to the notion of “equalized odds” proposed by Hardt et al. (2016). We study the problem of learning such a non-discriminatory predictor from a finite training set, both statistically and computationally. We show that a post-hoc correction approach, as suggested by Hardt et al, can be highly suboptimal, present a nearly-optimal statistical procedure, argue that the associated computational problem is intractable, and suggest a second moment relaxation of the non-discrimination definition for which learning is tractable.} }
Endnote
%0 Conference Paper %T Learning Non-Discriminatory Predictors %A Blake Woodworth %A Suriya Gunasekar %A Mesrob I. Ohannessian %A Nathan Srebro %B Proceedings of the 2017 Conference on Learning Theory %C Proceedings of Machine Learning Research %D 2017 %E Satyen Kale %E Ohad Shamir %F pmlr-v65-woodworth17a %I PMLR %P 1920--1953 %U https://proceedings.mlr.press/v65/woodworth17a.html %V 65 %X We consider learning a predictor which is non-discriminatory with respect to a “protected attribute” according to the notion of “equalized odds” proposed by Hardt et al. (2016). We study the problem of learning such a non-discriminatory predictor from a finite training set, both statistically and computationally. We show that a post-hoc correction approach, as suggested by Hardt et al, can be highly suboptimal, present a nearly-optimal statistical procedure, argue that the associated computational problem is intractable, and suggest a second moment relaxation of the non-discrimination definition for which learning is tractable.
APA
Woodworth, B., Gunasekar, S., Ohannessian, M.I. & Srebro, N.. (2017). Learning Non-Discriminatory Predictors. Proceedings of the 2017 Conference on Learning Theory, in Proceedings of Machine Learning Research 65:1920-1953 Available from https://proceedings.mlr.press/v65/woodworth17a.html.

Related Material