The learning component of dynamic allocation indices

DOI10.1214/aos/1176348788zbMath0760.62080OpenAlexW2063853624MaRDI QIDQ1206729

You-Gan Wang, J. C. Gittins

Publication date: 1 April 1993

Published in: The Annals of Statistics (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1214/aos/1176348788

zbMATH Keywords

Gittins indexexponential discountingnormal distributionsmultiarmed bandit problemBernoulli reward processdynamical allocation indexexpected immediate rewardlearning componentoptimal allocation ruletarget processesupward adjustment

Mathematics Subject Classification ID

Bayesian problems; characterization of Bayes procedures (62C10) Sequential statistical methods (62L99) Optimal stochastic control (93E20) Markov and semi-Markov decision processes (90C40) Sequential statistical design (62L05)

Related Items (9)

Bayesian adaptive bandit-based designs using the Gittins index for multi-armed trials with normally distributed endpoints ⋮ Generalized two-stage bandit problem ⋮ Bandit bounds from stochastic variability extrema ⋮ One-armed bandit process with a covariate ⋮ A confirmation of a conjecture on Feldman’s two-armed bandit problem ⋮ The prediction distribution for the heteroscedastic multivariate lineary models ⋮ Optimal learning with non-Gaussian rewards ⋮ Small-sample performance of Bernoulli two-armed bandit Bayesian strategies ⋮ Multi-armed bandit models for the optimal design of clinical trials: benefits and challenges

This page was built for publication: The learning component of dynamic allocation indices