Max K-armed bandit: On the ExtremeHunter algorithm and beyond
Machine Learning
2017-07-28 v1 Machine Learning
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 and Valko (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.
Keywords
Cite
@article{arxiv.1707.08820,
title = {Max K-armed bandit: On the ExtremeHunter algorithm and beyond},
author = {Mastane Achab and Stephan Clémençon and Aurélien Garivier and Anne Sabourin and Claire Vernade},
journal= {arXiv preprint arXiv:1707.08820},
year = {2017}
}