Passer à la navigation principale Passer à la recherche Passer au contenu principal

On Limited-Memory Subsampling Strategies for Bandits

Résultats de recherche: Le chapitre dans un livre, un rapport, une anthologie ou une collectionContribution à une conférenceRevue par des pairs

Résumé

There has been a recent surge of interest in nonparametric bandit algorithms based on subsampling. One drawback however of these approaches is the additional complexity required by random subsampling and the storage of the full history of rewards. Our first contribution is to show that a simple deterministic subsampling rule, proposed in the recent work of Baudry et al. (2020) under the name of “last-block subsampling”, is asymptotically optimal in one-parameter exponential families. In addition, we prove that these guarantees also hold when limiting the algorithm memory to a polylogarithmic function of the time horizon. These findings open up new perspectives, in particular for non-stationary scenarios in which the arm distributions evolve over time. We propose a variant of the algorithm in which only the most recent observations are used for subsampling, achieving optimal regret guarantees under the assumption of a known number of abrupt changes. Extensive numerical simulations highlight the merits of this approach, particularly when the changes are not only affecting the means of the rewards.

langue originaleAnglais
titreProceedings of the 38th International Conference on Machine Learning, ICML 2021
EditeurML Research Press
Pages727-737
Nombre de pages11
ISBN (Electronique)9781713845065
étatPublié - 1 janv. 2021
Modification externeOui
Evénement38th International Conference on Machine Learning, ICML 2021 - Virtual, Online
Durée: 18 juil. 202124 juil. 2021

Série de publications

NomProceedings of Machine Learning Research
Volume139
ISSN (Electronique)2640-3498

Une conférence

Une conférence38th International Conference on Machine Learning, ICML 2021
La villeVirtual, Online
période18/07/2124/07/21

Empreinte digitale

Examiner les sujets de recherche de « On Limited-Memory Subsampling Strategies for Bandits ». Ensemble, ils forment une empreinte digitale unique.

Contient cette citation