From the archives

Max K-armed bandits: On the ExtremeHunter algorithm and beyond

Historical material. Affiliations, schedules and roles on this page describe their original context.

ConferenceName:  European Conference on Machine Learning

url:  ECML-PKDD 2017

Edition Number:  2 017

Date:  September, 2017

Place:  Skopje, Macedonia

Authors:  Mastane AchabStéphan ClémençonAurélien GarivierAnne SabourinClaire Vernade

Abstract: 

This paper is devoted to the study of the max K-armed bandit problem, which consists in sequentially allocating resources in order to detect extreme values. Our contribution is twofold.
We first significantly refine the analysis of the ExtremeHunter algorithm carried out in [Carpentier et al. 2014], and next propose an alternative approach, showing that, remarkably, Extreme Bandits can be reduced to a classical version of the bandit problem to a certain extent. Beyond the formal analysis, these two approaches are compared through numerical experiments.

Direct link:  published in

Keywords: