Saved in:
Bibliographic Details
Main Authors: Soma, Tasuku, Gatmiry, Khashayar, Gupta, Sharut, Jegelka, Stefanie
Format: Preprint
Published: 2022
Subjects:
Online Access:https://arxiv.org/abs/2212.13669
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915129916915712
author Soma, Tasuku
Gatmiry, Khashayar
Gupta, Sharut
Jegelka, Stefanie
author_facet Soma, Tasuku
Gatmiry, Khashayar
Gupta, Sharut
Jegelka, Stefanie
contents Distributionally robust optimization (DRO) can improve the robustness and fairness of learning methods. In this paper, we devise stochastic algorithms for a class of DRO problems including group DRO, subpopulation fairness, and empirical conditional value at risk (CVaR) optimization. Our new algorithms achieve faster convergence rates than existing algorithms for multiple DRO settings. We also provide a new information-theoretic lower bound that implies our bounds are tight for group DRO. Empirically, too, our algorithms outperform known methods.
format Preprint
id arxiv_https___arxiv_org_abs_2212_13669
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Near-Optimal Algorithms for Group Distributionally Robust Optimization and Beyond
Soma, Tasuku
Gatmiry, Khashayar
Gupta, Sharut
Jegelka, Stefanie
Machine Learning
Optimization and Control
Distributionally robust optimization (DRO) can improve the robustness and fairness of learning methods. In this paper, we devise stochastic algorithms for a class of DRO problems including group DRO, subpopulation fairness, and empirical conditional value at risk (CVaR) optimization. Our new algorithms achieve faster convergence rates than existing algorithms for multiple DRO settings. We also provide a new information-theoretic lower bound that implies our bounds are tight for group DRO. Empirically, too, our algorithms outperform known methods.
title Near-Optimal Algorithms for Group Distributionally Robust Optimization and Beyond
topic Machine Learning
Optimization and Control
url https://arxiv.org/abs/2212.13669