An $\varepsilon$-Best-Arm Identification Algorithm for Fixed-Confidence and Beyond
Abstract
We propose EB-TC, a novel sampling rule for -best arm identification in stochastic bandits. It is the first instance of Top Two algorithm analyzed for approximate best arm identification. EB-TC is an *anytime* sampling rule that can therefore be employed without modification for fixed confidence or fixed budget identification (without prior knowledge of the budget). We provide three types of theoretical guarantees for EB-TC. First, we prove bounds on its expected sample complexity in the fixed confidence setting, notably showing its asymptotic optimality in combination with an adaptive tuning of its exploration parameter. We complement these findings with upper bounds on its probability of error at any time and for any error parameter, which further yield upper bounds on its simple regret at any time. Finally, we show through numerical simulations that EB-TC performs favorably compared to existing algorithms, in different settings.
Cite
@article{arxiv.2305.16041,
title = {An $\varepsilon$-Best-Arm Identification Algorithm for Fixed-Confidence and Beyond},
author = {Marc Jourdan and Rémy Degenne and Emilie Kaufmann},
journal= {arXiv preprint arXiv:2305.16041},
year = {2023}
}
Comments
68 pages, 14 figures, 4 tables. To be published in the Thirty-seventh Conference on Neural Information Processing Systems