Approximate information for efficient exploration-exploitation strategies - Archive ouverte HAL Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2023

Approximate information for efficient exploration-exploitation strategies

Information approximée pour des stratégies d'exploration-exploitation efficaces

Résumé

This paper addresses the exploration-exploitation dilemma inherent in decision-making, focusing on multi-armed bandit problems. The problems involve an agent deciding whether to exploit current knowledge for immediate gains or explore new avenues for potential long-term rewards. We here introduce a novel algorithm, approximate information maximization (AIM), which employs an analytical approximation of the entropy gradient to choose which arm to pull at each point in time. AIM matches the performance of Infomax and Thompson sampling while also offering enhanced computational speed, determinism, and tractability. Empirical evaluation of AIM indicates its compliance with the Lai & Robbins asymptotic bound and demonstrates its robustness for a range of priors. Its expression is tunable, which allows for specific optimization in various settings.
Fichier principal
Vignette du fichier
main.pdf (701.24 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Licence : CC BY - Paternité

Dates et versions

hal-04147006 , version 1 (03-07-2023)

Licence

Paternité

Identifiants

  • HAL Id : hal-04147006 , version 1

Citer

Alex Barbier--Chebbah, Christian L. Vestergaard, Jean-Baptiste Masson. Approximate information for efficient exploration-exploitation strategies. 2023. ⟨hal-04147006⟩
14 Consultations
9 Téléchargements

Partager

Gmail Facebook X LinkedIn More