Fixed-confidence guarantees for Bayesian best-arm identification

Shang, Xuedong; de Heide, Rianne; Kaufmann, Emilie; Ménard, Pierre; Valko, Michal

Type de document :

Communication dans un congrès avec actes

Titre :

Fixed-confidence guarantees for Bayesian best-arm identification

Auteur(s) :

Shang, Xuedong [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Sequential Learning [SEQUEL]
Scool [Scool]
de Heide, Rianne [Auteur]
Universiteit Leiden = Leiden University
Centrum Wiskunde & Informatica [CWI]
Kaufmann, Emilie [Auteur]

Sequential Learning [SEQUEL]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Centre National de la Recherche Scientifique [CNRS]
Scool [Scool]
Ménard, Pierre [Auteur]
Sequential Learning [SEQUEL]
Scool [Scool]
Valko, Michal [Auteur]

DeepMind [Paris]

Titre de la manifestation scientifique :

International Conference on Artificial Intelligence and Statistics

Ville :

Palermo

Pays :

Italie

Date de début de la manifestation scientifique :

2020

Discipline(s) HAL :

Informatique [cs]/Apprentissage [cs.LG]

Résumé en anglais : [en]

We investigate and provide new insights on the sampling rule called Top-Two Thompson Sampling (TTTS). In particular, we justify its use for fixed-confidence best-arm identification. We further propose a variant of TTTS ...
Lire la suite >We investigate and provide new insights on the sampling rule called Top-Two Thompson Sampling (TTTS). In particular, we justify its use for fixed-confidence best-arm identification. We further propose a variant of TTTS called Top-Two Transportation Cost (T3C), which disposes of the computational burden of TTTS. As our main contribution, we provide the first sample complexity analysis of TTTS and T3C when coupled with a very natural Bayesian stopping rule, for bandits with Gaussian rewards, solving one of the open questions raised by Russo (2016). We also provide new posterior convergence results for TTTS under two models that are commonly used in practice: bandits with Gaussian and Bernoulli rewards and conjugate priors.Lire moins >

Langue :

Anglais

Comité de lecture :

Oui

Audience :

Internationale

Vulgarisation :

Non

Projet ANR :

BANDITS MANCHOTS POUR SIGNAUX NON-STATIONNAIRES ET STRUCTURES
Au delà de l'apprentissage séquentiel pour de meilleures prises de décisions

Commentaire :

Due to COVID-19, the event has been postponed. AISTATS 2020 held online on August 26-28th.