Title :

Pure Exploration in Bandits with Linear Constraints

Author(s) :

Carlsson, Emil [Auteur]
Chalmers University of Technology [Gothenburg, Sweden]
Basu, Debabrota [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Scool [Scool]
Johansson, Fredrik D. [Auteur]
Chalmers University of Technology [Gothenburg, Sweden]
Dubhashi, Devdatt [Auteur]
Chalmers University of Technology [Gothenburg, Sweden]

Conference title :

International Conference on Artificial Intelligence and Statistics

City :

Valencia (Espagne)

Country :

Espagne

Start date of the conference :

2024-05

Book title :

Proceedings of Machine Learning Research (PMLR)

Journal title :

Proceedings of Machine Learning Research (PMLR)

Publication date :

2024-05

HAL domain(s) :

Informatique [cs]/Apprentissage [cs.LG]
Informatique [cs]/Intelligence artificielle [cs.AI]
Informatique [cs]/Théorie de l'information [cs.IT]
Statistiques [stat]/Machine Learning [stat.ML]

English abstract : [en]

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 ...
Show more >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.Show less >

Language :

Anglais

Peer reviewed article :

Oui

Audience :

Internationale

Popular science :

Non

ANR Project :

REPUBLIC: Vers l'IA responsable avec l'apprentissage par renforcement sous contraintes
Causal eXplanations in Reinforcement Learning

Collections :

• Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189

Source :

Harvested from HAL