FAIRM: Learning invariant representations for algorithmic fairness and domain generalization with minimax optimality

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Sai, Zhang, Linjun
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929300087767040
author Li, Sai
Zhang, Linjun
author_facet Li, Sai
Zhang, Linjun
contents Machine learning methods often assume that the test data have the same distribution as the training data. However, this assumption may not hold due to multiple levels of heterogeneity in applications, raising issues in algorithmic fairness and domain generalization. In this work, we address the problem of fair and generalizable machine learning by invariant principles. We propose a training environment-based oracle, FAIRM, which has desirable fairness and domain generalization properties under a diversity-type condition. We then provide an empirical FAIRM with finite-sample theoretical guarantees under weak distributional assumptions. We then develop efficient algorithms to realize FAIRM in linear models and demonstrate the nonasymptotic performance with minimax optimality. We evaluate our method in numerical experiments with synthetic data and MNIST data and show that it outperforms its counterparts.
format Preprint
id arxiv_https___arxiv_org_abs_2404_01608
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle FAIRM: Learning invariant representations for algorithmic fairness and domain generalization with minimax optimality
Li, Sai
Zhang, Linjun
Machine Learning
Methodology
Machine learning methods often assume that the test data have the same distribution as the training data. However, this assumption may not hold due to multiple levels of heterogeneity in applications, raising issues in algorithmic fairness and domain generalization. In this work, we address the problem of fair and generalizable machine learning by invariant principles. We propose a training environment-based oracle, FAIRM, which has desirable fairness and domain generalization properties under a diversity-type condition. We then provide an empirical FAIRM with finite-sample theoretical guarantees under weak distributional assumptions. We then develop efficient algorithms to realize FAIRM in linear models and demonstrate the nonasymptotic performance with minimax optimality. We evaluate our method in numerical experiments with synthetic data and MNIST data and show that it outperforms its counterparts.
title FAIRM: Learning invariant representations for algorithmic fairness and domain generalization with minimax optimality
topic Machine Learning
Methodology
url https://arxiv.org/abs/2404.01608