Saved in:
Bibliographic Details
Main Authors: Sarkar, Dhruv, Pandey, Nishant, Chowdhury, Sayak Ray
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.22969
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908800373489664
author Sarkar, Dhruv
Pandey, Nishant
Chowdhury, Sayak Ray
author_facet Sarkar, Dhruv
Pandey, Nishant
Chowdhury, Sayak Ray
contents Nash regret has recently emerged as a principled fairness-aware performance metric for stochastic multi-armed bandits, motivated by the Nash Social Welfare objective. Although this notion has been extended to linear bandits, existing results suffer from suboptimality in ambient dimension $d$, stemming from proof techniques that rely on restrictive concentration inequalities. In this work, we resolve this open problem by introducing new analytical tools that yield an order-optimal Nash regret bound in linear bandits. Beyond Nash regret, we initiate the study of $p$-means regret in linear bandits, a unifying framework that interpolates between fairness and utility objectives and strictly generalizes Nash regret. We propose a generic algorithmic framework, FairLinBandit, that works as a meta-algorithm on top of any linear bandit strategy. We instantiate this framework using two bandit algorithms: Phased Elimination and Upper Confidence Bound, and prove that both achieve sublinear $p$-means regret for the entire range of $p$. Extensive experiments on linear bandit instances generated from real-world datasets demonstrate that our methods consistently outperform the existing state-of-the-art baseline.
format Preprint
id arxiv_https___arxiv_org_abs_2601_22969
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Improved Algorithms for Nash Welfare in Linear Bandits
Sarkar, Dhruv
Pandey, Nishant
Chowdhury, Sayak Ray
Machine Learning
Nash regret has recently emerged as a principled fairness-aware performance metric for stochastic multi-armed bandits, motivated by the Nash Social Welfare objective. Although this notion has been extended to linear bandits, existing results suffer from suboptimality in ambient dimension $d$, stemming from proof techniques that rely on restrictive concentration inequalities. In this work, we resolve this open problem by introducing new analytical tools that yield an order-optimal Nash regret bound in linear bandits. Beyond Nash regret, we initiate the study of $p$-means regret in linear bandits, a unifying framework that interpolates between fairness and utility objectives and strictly generalizes Nash regret. We propose a generic algorithmic framework, FairLinBandit, that works as a meta-algorithm on top of any linear bandit strategy. We instantiate this framework using two bandit algorithms: Phased Elimination and Upper Confidence Bound, and prove that both achieve sublinear $p$-means regret for the entire range of $p$. Extensive experiments on linear bandit instances generated from real-world datasets demonstrate that our methods consistently outperform the existing state-of-the-art baseline.
title Improved Algorithms for Nash Welfare in Linear Bandits
topic Machine Learning
url https://arxiv.org/abs/2601.22969