Answers are grounded in this paper and cite their sources.
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.
No takes yet. Share an insight, caveat, or question.
Peter W. Jones (1976) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: