On Limited-Memory Subsampling Strategies for Bandits - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2021

On Limited-Memory Subsampling Strategies for Bandits

Abstract

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.
Fichier principal
Vignette du fichier
main.pdf (1.24 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03265442 , version 1 (20-06-2021)

Identifiers

Cite

Dorian Baudry, Yoan Russac, Olivier Cappé. On Limited-Memory Subsampling Strategies for Bandits. ICML 2021- International Conference on Machine Learning, Jul 2021, Vienna / Virtual, Austria. ⟨hal-03265442⟩
65 View
53 Download

Altmetric

Share

Gmail Facebook X LinkedIn More