An asymptotically optimal policy for finite support models in the multiarmed bandit problem

DOI10.1007/S10994-011-5257-4zbMath1237.91037arXiv0905.2776OpenAlexW2131958277WikidataQ56675674 ScholiaQ56675674MaRDI QIDQ415624

Junya Honda, Akimichi Takemura

Publication date: 8 May 2012

Published in: Machine Learning (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/0905.2776

zbMATH Keywords

convex optimizationbandit problemsfinite-time regretMED policy

Mathematics Subject Classification ID

Convex programming (90C25) Stochastic games, stochastic differential games (91A15) Rationality and learning in game theory (91A26)

Related Items (9)

Infomax strategies for an optimal balance between exploration and exploitation ⋮ EXPLORATION–EXPLOITATION POLICIES WITH ALMOST SURE, ARBITRARILY SLOW GROWING ASYMPTOTIC REGRET ⋮ Kullback-Leibler upper confidence bounds for optimal sequential allocation ⋮ Unnamed Item ⋮ A perpetual search for talents across overlapping generations: a learning process ⋮ ASYMPTOTICALLY OPTIMAL MULTI-ARMED BANDIT POLICIES UNDER A COST CONSTRAINT ⋮ On Bayesian index policies for sequential resource allocation ⋮ Unnamed Item ⋮ Unnamed Item

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Introduction to sensitivity and stability analysis in nonlinear programming
• Nonparametric bandit methods
• Asymptotically efficient adaptive allocation rules
• Multi-armed bandit problem revisited
• Optimal adaptive policies for sequential allocation problems
• Exploration of multi-state environments: Local measures and back-propagation of uncertainty
• The Multi-Armed Bandit Problem: Decomposition and Computation
• Non-overlapping domain decomposition for evolution operators
• The Nonstochastic Multiarmed Bandit Problem
• Sample mean based index policies by O(log n) regret for the multi-armed bandit problem
• The Continuum-Armed Bandit Problem
• Elements of Information Theory
• Some aspects of the sequential design of experiments
• Convergence of stochastic processes
• Finite-time analysis of the multiarmed bandit problem

This page was built for publication: An asymptotically optimal policy for finite support models in the multiarmed bandit problem