Computing a Classic Index for Finite-Horizon Bandits

DOI10.1287/IJOC.1100.0398zbMath1243.90157OpenAlexW2154087138MaRDI QIDQ2899118

José Niño-Mora

Publication date: 28 July 2012

Published in: INFORMS Journal on Computing (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1287/ijoc.1100.0398

zbMATH Keywords

dynamic programmingfinite-horizon indexMarkov bandits

Mathematics Subject Classification ID

Stochastic programming (90C15) Dynamic programming (90C39) Stopping times; optimal stopping problems; gambling theory (60G40) Markov and semi-Markov decision processes (90C40) Statistical methods; economic indices and measures (91B82) Probabilistic games; gambling (91A60)

Related Items (9)

Generalized Restless Bandits and the Knapsack Problem for Perishable Inventories ⋮ Bayesian policy reuse ⋮ BANDIT STRATEGIES EVALUATED IN THE CONTEXT OF CLINICAL TRIALS IN RARE LIFE-THREATENING DISEASES ⋮ On Bayesian index policies for sequential resource allocation ⋮ Coupled bisection for root ordering ⋮ An asymptotically optimal heuristic for general nonstationary finite-horizon restless multi-armed, multi-action bandits ⋮ Learning to Optimize via Information-Directed Sampling ⋮ Technical Note—A Note on the Equivalence of Upper Confidence Bounds and Gittins Indices for Patient Agents ⋮ Multi-armed bandit models for the optimal design of clinical trials: benefits and challenges

This page was built for publication: Computing a Classic Index for Finite-Horizon Bandits