Simple regret for infinitely many armed bandits - Archive ouverte HAL Access content directly
Conference Papers Year :

Simple regret for infinitely many armed bandits

(1) , (2)
1
2
Alexandra Carpentier
  • Function : Author
  • PersonId : 961172
Michal Valko

Abstract

We consider a stochastic bandit problem with infinitely many arms. In this setting, the learner has no chance of trying all the arms even once and has to dedicate its limited number of samples only to a certain number of arms. All previous algorithms for this setting were designed for minimizing the cumulative regret of the learner. In this paper, we propose an algorithm aiming at minimizing the simple regret. As in the cumulative regret setting of infinitely many armed bandits , the rate of the simple regret will depend on a parameter β characterizing the distribution of the near-optimal arms. We prove that depending on β, our algorithm is minimax optimal either up to a multiplicative constant or up to a log(n) factor. We also provide extensions to several important cases: when β is unknown, in a natural setting where the near-optimal arms have a small variance , and in the case of unknown time horizon.
Fichier principal
Vignette du fichier
carpentier2015simple.pdf (402.4 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01153538 , version 1 (19-05-2015)

Identifiers

Cite

Alexandra Carpentier, Michal Valko. Simple regret for infinitely many armed bandits. International Conference on Machine Learning, Jul 2015, Lille, France. ⟨hal-01153538⟩
169 View
119 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More