Adaptive policies for sequential sampling under incomplete information and a cost constraint

DOI10.1007/978-1-4614-4109-0_8zbMATH Open1315.93091arXiv1201.4002OpenAlexW1898012344MaRDI QIDQ5261007FDOQ5261007

Authors: Odysseas Kanavetas, A. N. Burnetas 

Publication date: 1 July 2015

Published in: Applications of Mathematics and Informatics in Military Science (Search for Journal in Brave)

Abstract: We consider the problem of sequential sampling from a finite number of independent statistical populations to maximize the expected infinite horizon average outcome per period, under a constraint that the expected average sampling cost does not exceed an upper bound. The outcome distributions are not known. We construct a class of consistent adaptive policies, under which the average outcome converges with probability 1 to the true value under complete information for all distributions with finite means. We also compare the rate of convergence for various policies in this class using simulation.

Full work available at URL: https://arxiv.org/abs/1201.4002

Recommendations

• Optimal adaptive policies for sequential allocation problems
• Obtaining the best value for money in adaptive sequential estimation
• Asymptotically optimal multi-armed bandit policies under a cost constraint
• Consistency of sequential Bayesian sampling policies
• Optimal sequential sampling from two populations.

zbMATH Keywords

sequential designstochastic learning and adaptive controlsampling cost constraint

Mathematics Subject Classification ID

Sequential statistical design (62L05) Stochastic learning and adaptive control (93E35)

Cites Work

• Asymptotically efficient adaptive allocation rules
• Some aspects of the sequential design of experiments
• Finite-time analysis of the multiarmed bandit problem
• Optimal Adaptive Policies for Markov Decision Processes
• Optimal adaptive policies for sequential allocation problems
• Finite-time lower bounds for the two-armed bandit problem
• Learning Theory
• Gittins indices and constrained allocation in clinical trials

Cited In (3)

• Asymptotically optimal multi-armed bandit policies under a cost constraint
• Sequential sampling to myopically maximize the expected value of information
• Consistency of sequential Bayesian sampling policies

This page was built for publication: Adaptive policies for sequential sampling under incomplete information and a cost constraint

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5261007)