Minimal Exploration in Structured Stochastic Bandits - CentraleSupélec Accéder directement au contenu
Communication Dans Un Congrès Année : 2017

Minimal Exploration in Structured Stochastic Bandits

Résumé

This paper introduces and addresses a wide class of stochastic bandit problems where the function mapping the arm to the corresponding reward exhibits some known structural properties. Most existing structures (e.g. linear, Lipschitz, unimodal, combinatorial, dueling, . . . ) are covered by our framework. We derive an asymptotic instance-specific regret lower bound for these problems, and develop OSSB, an algorithm whose regret matches this fundamental limit. OSSB is not based on the classical principle of "optimism in the face of uncertainty" or on Thompson sampling, and rather aims at matching the minimal exploration rates of sub-optimal arms as characterized in the derivation of the regret lower bound. We illustrate the efficiency of OSSB using numerical experiments in the case of the linear bandit problem and show that OSSB outperforms existing algorithms, including Thompson sampling.
Fichier principal
Vignette du fichier
combes_minimal_exploration_nips2017.pdf (453.94 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02395029 , version 1 (09-04-2020)

Identifiants

Citer

Richard Combes, Stefan Magureanu, Alexandre Proutiere. Minimal Exploration in Structured Stochastic Bandits. 31st Conference on Neural Information Processing Systems (NIPS), Dec 2017, Long Beach, CA, United States. ⟨hal-02395029⟩
46 Consultations
23 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More