Trading off rewards and errors in multi-armed bandits
Machine Learning
2026-05-04 v1
Abstract
In multi-armed bandits, the most-explored arms are the most informative, while reward maximization typically pulls only the best arm. We study the tradeoff between identifying arm means accurately and accumulating reward, and present an algorithm with regret guarantees that interpolates between the two objectives. We provide both upper and lower bounds and validate empirically.
Keywords
Cite
@article{arxiv.2605.00488,
title = {Trading off rewards and errors in multi-armed bandits},
author = {Akram Erraqabi and Alessandro Lazaric and Michal Valko and Emma Brunskill and Yun-En Liu},
journal= {arXiv preprint arXiv:2605.00488},
year = {2026}
}
Comments
Published at AISTATS 2017 (20th International Conference on Artificial Intelligence and Statistics)