Some results for the two armed bandit problem
- 1 January 1976
- journal article
- research article
- Published by Taylor & Francis in Mathematische Operationsforschung und Statistik
- Vol. 7 (3) , 471-475
- https://doi.org/10.1080/02331887608801311
Abstract
The paper is concerned with the optimal dynamic programming approach to the solution of the two armed bandit problem for beta priors for the two unknown probabilities. Some properties of the objctive function are obtained and a conjecture concerning the design is made. The suboptimal one step ahead design in considered and is shown to have the same properties as those of the optimal only in certain special cases.Keywords
This publication has 1 reference indexed in Scilit:
- The two-armed banditBiometrika, 1975