QuACK: A Multipurpose Queuing Algorithm for Cooperative $k$-Armed Bandits

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Howson, Benjamin, Filippi, Sarah, Pike-Burke, Ciara
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866929570124398592
author Howson, Benjamin
Filippi, Sarah
Pike-Burke, Ciara
author_facet Howson, Benjamin
Filippi, Sarah
Pike-Burke, Ciara
contents We study the cooperative stochastic $k$-armed bandit problem, where a network of $m$ agents collaborate to find the optimal action. In contrast to most prior work on this problem, which focuses on extending a specific algorithm to the multi-agent setting, we provide a black-box reduction that allows us to extend any single-agent bandit algorithm to the multi-agent setting. Under mild assumptions on the bandit environment, we prove that our reduction transfers the regret guarantees of the single-agent algorithm to the multi-agent setting. These guarantees are tight in subgaussian environments, in that using a near minimax optimal single-player algorithm is near minimax optimal in the multi-player setting up to an additive graph-dependent quantity. Our reduction and theoretical results are also general, and apply to many different bandit settings. By plugging in appropriate single-player algorithms, we can easily develop provably efficient algorithms for many multi-player settings such as heavy-tailed bandits, duelling bandits and bandits with local differential privacy, among others. Experimentally, our approach is competitive with or outperforms specialised multi-agent algorithms.
format Preprint
id arxiv_https___arxiv_org_abs_2410_23867
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle QuACK: A Multipurpose Queuing Algorithm for Cooperative $k$-Armed Bandits
Howson, Benjamin
Filippi, Sarah
Pike-Burke, Ciara
Machine Learning
We study the cooperative stochastic $k$-armed bandit problem, where a network of $m$ agents collaborate to find the optimal action. In contrast to most prior work on this problem, which focuses on extending a specific algorithm to the multi-agent setting, we provide a black-box reduction that allows us to extend any single-agent bandit algorithm to the multi-agent setting. Under mild assumptions on the bandit environment, we prove that our reduction transfers the regret guarantees of the single-agent algorithm to the multi-agent setting. These guarantees are tight in subgaussian environments, in that using a near minimax optimal single-player algorithm is near minimax optimal in the multi-player setting up to an additive graph-dependent quantity. Our reduction and theoretical results are also general, and apply to many different bandit settings. By plugging in appropriate single-player algorithms, we can easily develop provably efficient algorithms for many multi-player settings such as heavy-tailed bandits, duelling bandits and bandits with local differential privacy, among others. Experimentally, our approach is competitive with or outperforms specialised multi-agent algorithms.
title QuACK: A Multipurpose Queuing Algorithm for Cooperative $k$-Armed Bandits
topic Machine Learning
url https://arxiv.org/abs/2410.23867