Beyond Minimax Rates in Group Distributionally Robust Optimization via a Novel Notion of Sparsity

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Nguyen, Quan, Mehta, Nishant A., Guzmán, Cristóbal
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913672241086464
author Nguyen, Quan
Mehta, Nishant A.
Guzmán, Cristóbal
author_facet Nguyen, Quan
Mehta, Nishant A.
Guzmán, Cristóbal
contents The minimax sample complexity of group distributionally robust optimization (GDRO) has been determined up to a $\log(K)$ factor, where $K$ is the number of groups. In this work, we venture beyond the minimax perspective via a novel notion of sparsity that we dub $(λ, β)$-sparsity. In short, this condition means that at any parameter $θ$, there is a set of at most $β$ groups whose risks at $θ$ all are at least $λ$ larger than the risks of the other groups. To find an $ε$-optimal $θ$, we show via a novel algorithm and analysis that the $ε$-dependent term in the sample complexity can swap a linear dependence on $K$ for a linear dependence on the potentially much smaller $β$. This improvement leverages recent progress in sleeping bandits, showing a fundamental connection between the two-player zero-sum game optimization framework for GDRO and per-action regret bounds in sleeping bandits. We next show an adaptive algorithm which, up to log factors, gets a sample complexity bound that adapts to the best $(λ, β)$-sparsity condition that holds. We also show how to get a dimension-free semi-adaptive sample complexity bound with a computationally efficient method. Finally, we demonstrate the practicality of the $(λ, β)$-sparsity condition and the improved sample efficiency of our algorithms on both synthetic and real-life datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2410_00690
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Beyond Minimax Rates in Group Distributionally Robust Optimization via a Novel Notion of Sparsity
Nguyen, Quan
Mehta, Nishant A.
Guzmán, Cristóbal
Machine Learning
Artificial Intelligence
Optimization and Control
The minimax sample complexity of group distributionally robust optimization (GDRO) has been determined up to a $\log(K)$ factor, where $K$ is the number of groups. In this work, we venture beyond the minimax perspective via a novel notion of sparsity that we dub $(λ, β)$-sparsity. In short, this condition means that at any parameter $θ$, there is a set of at most $β$ groups whose risks at $θ$ all are at least $λ$ larger than the risks of the other groups. To find an $ε$-optimal $θ$, we show via a novel algorithm and analysis that the $ε$-dependent term in the sample complexity can swap a linear dependence on $K$ for a linear dependence on the potentially much smaller $β$. This improvement leverages recent progress in sleeping bandits, showing a fundamental connection between the two-player zero-sum game optimization framework for GDRO and per-action regret bounds in sleeping bandits. We next show an adaptive algorithm which, up to log factors, gets a sample complexity bound that adapts to the best $(λ, β)$-sparsity condition that holds. We also show how to get a dimension-free semi-adaptive sample complexity bound with a computationally efficient method. Finally, we demonstrate the practicality of the $(λ, β)$-sparsity condition and the improved sample efficiency of our algorithms on both synthetic and real-life datasets.
title Beyond Minimax Rates in Group Distributionally Robust Optimization via a Novel Notion of Sparsity
topic Machine Learning
Artificial Intelligence
Optimization and Control
url https://arxiv.org/abs/2410.00690