2 problems

0 votes0 replies0 views
Berry's conjecture on the dynamic-programming design's asymptotic failure-count criterion
Consider the finite-horizon Bayesian two-armed bandit and the deterministic dynamic-programming design, denoted by $\operatorname{DP}$ . Let $t\ll T$ denote epochs with a large numb…
Conjectureopenbayesian-decision-theory multi-armed-banditsarXiv posted 2 days ago
0 votes0 replies0 views
Bradt, Johnson and Karlin's optimality conjecture for the noncomplementary two-arm bandit
Let the two arms have success probabilities from the known set $\{\theta_C,\theta_D\}$ , where $\theta_C+\theta_D\neq 1$ , with the assignment of probabilities to arms unknown. A des…
Conjectureopenbayesian-decision-theory multi-armed-banditsarXiv posted 2 days ago

Berry's conjecture on the dynamic-programming design's asymptotic failure-count criterion