Qualitative Multi-Armed Bandits: A Quantile-Based Approach

Abstract : We formalize and study the multi-armed bandit (MAB) problem in a generalized stochastic setting, in which rewards are not assumed to be numerical. Instead, rewards are measured on a qualitative scale that allows for comparison but invalidates arithmetic operations such as averaging. Correspondingly, instead of characterizing an arm in terms of the mean of the underlying distribution, we opt for using a quantile of that distribution as a representative value. We address the problem of quantile-based online learning both for the case of a finite (pure exploration) and infinite time horizon (cumulative regret minimization). For both cases, we propose suitable algorithms and analyze their properties. These properties are also illustrated by means of first experimental studies.
Complete list of metadatas

Cited literature [25 references]  Display  Hide  Download

https://hal.inria.fr/hal-01204708
Contributor : Balazs Szorenyi <>
Submitted on : Thursday, September 24, 2015 - 2:36:36 PM
Last modification on : Friday, March 22, 2019 - 1:35:34 AM
Long-term archiving on : Tuesday, December 29, 2015 - 9:49:32 AM

File

qmab_final.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01204708, version 1

Citation

Balazs Szorenyi, Róbert Busa-Fekete, Paul Weng, Eyke Hüllermeier. Qualitative Multi-Armed Bandits: A Quantile-Based Approach. 32nd International Conference on Machine Learning, Jul 2015, Lille, France. pp.1660-1668. ⟨hal-01204708⟩

Share

Metrics

Record views

1064

Files downloads

411