Pure Exploration in Bandits with Linear Constraints - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Pure Exploration in Bandits with Linear Constraints

Résumé

We address the problem of identifying the optimal policy with a fixed confidence level in a multi-armed bandit setup, when the arms are subject to linear constraints. Unlike the standard best-arm identification problem which is well studied, the optimal policy in this case may not be deterministic and could mix between several arms. This changes the geometry of the problem which we characterize via an information-theoretic lower bound. We introduce two asymptotically optimal algorithms for this setting, one based on the Track-and-Stop method and the other based on a game-theoretic approach. Both these algorithms try to track an optimal allocation based on the lower bound and computed by a weighted projection onto the boundary of a normal cone. Finally, we provide empirical results that validate our bounds and visualize how constraints change the hardness of the problem.

Dates et versions

hal-04203235 , version 1 (11-09-2023)

Licence

Paternité - Pas d'utilisation commerciale

Identifiants

Citer

Emil Carlsson, Debabrota Basu, Fredrik D. Johansson, Devdatt Dubhashi. Pure Exploration in Bandits with Linear Constraints. EWRL 2023 – European Workshop on Reinforcement Learning, Sep 2023, Brussels, Belgium. ⟨hal-04203235⟩
34 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More