Multi-Bandit Best Arm Identification

Victor Gabillon 1 Mohammad Ghavamzadeh 1 Alessandro Lazaric 1 Sébastien Bubeck 2
1 SEQUEL - Sequential Learning
LIFL - Laboratoire d'Informatique Fondamentale de Lille, LAGIS - Laboratoire d'Automatique, Génie Informatique et Signal, Inria Lille - Nord Europe
Abstract : We study the problem of identifying the best arm in each of the bandits in a multi-bandit multi-armed setting. We first propose an algorithm called Gap-based Exploration (GapE) that focuses on the arms whose mean is close to the mean of the best arm in the same bandit (i.e., small gap). We then introduce an algorithm, called GapE-V, which takes into account the variance of the arms in addition to their gap. We prove an upper-bound on the probability of error for both algorithms. Since GapE and GapE-V need to tune an exploration parameter that depends on the complexity of the problem, which is often unknown in advance, we also introduce variation of these algorithms that estimates this complexity online. Finally, we evaluate the performance of these algorithms and compare them to other allocation strategies on a number of synthetic problems.
Document type :
Reports
Complete list of metadatas

Cited literature [13 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-00632523
Contributor : Victor Gabillon <>
Submitted on : Saturday, November 19, 2011 - 3:11:40 PM
Last modification on : Thursday, February 21, 2019 - 10:52:49 AM
Long-term archiving on : Monday, February 20, 2012 - 2:20:57 AM

File

multi-bandit_techreport.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00632523, version 3

Collections

Citation

Victor Gabillon, Mohammad Ghavamzadeh, Alessandro Lazaric, Sébastien Bubeck. Multi-Bandit Best Arm Identification. 2011. ⟨hal-00632523v3⟩

Share

Metrics

Record views

389

Files downloads

184