Multi-Player Approaches for Dueling Bandits

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Raveh, Or, Honda, Junya, Sugiyama, Masashi
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915253563949056
author Raveh, Or
Honda, Junya
Sugiyama, Masashi
author_facet Raveh, Or
Honda, Junya
Sugiyama, Masashi
contents Various approaches have emerged for multi-armed bandits in distributed systems. The multiplayer dueling bandit problem, common in scenarios with only preference-based information like human feedback, introduces challenges related to controlling collaborative exploration of non-informative arm pairs, but has received little attention. To fill this gap, we demonstrate that the direct use of a Follow Your Leader black-box approach matches the lower bound for this setting when utilizing known dueling bandit algorithms as a foundation. Additionally, we analyze a message-passing fully distributed approach with a novel Condorcet-winner recommendation protocol, resulting in expedited exploration in many cases. Our experimental comparisons reveal that our multiplayer algorithms surpass single-player benchmark algorithms, underscoring their efficacy in addressing the nuanced challenges of the multiplayer dueling bandit setting.
format Preprint
id arxiv_https___arxiv_org_abs_2405_16168
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Multi-Player Approaches for Dueling Bandits
Raveh, Or
Honda, Junya
Sugiyama, Masashi
Machine Learning
Various approaches have emerged for multi-armed bandits in distributed systems. The multiplayer dueling bandit problem, common in scenarios with only preference-based information like human feedback, introduces challenges related to controlling collaborative exploration of non-informative arm pairs, but has received little attention. To fill this gap, we demonstrate that the direct use of a Follow Your Leader black-box approach matches the lower bound for this setting when utilizing known dueling bandit algorithms as a foundation. Additionally, we analyze a message-passing fully distributed approach with a novel Condorcet-winner recommendation protocol, resulting in expedited exploration in many cases. Our experimental comparisons reveal that our multiplayer algorithms surpass single-player benchmark algorithms, underscoring their efficacy in addressing the nuanced challenges of the multiplayer dueling bandit setting.
title Multi-Player Approaches for Dueling Bandits
topic Machine Learning
url https://arxiv.org/abs/2405.16168