4 problems

Let $\varepsilon>0$ denote the distance of the arrival rate to the capacity region, and suppose that $\varepsilon$ is unknown to the algorithm. Consider algorithms for the single-s…

LD-HOOT is a Monte Carlo tree search algorithm that uses LD-HOO to select actions in continuous action spaces. The goal is consistency in the sense that, as the number of samples t…

Adaptive exploration conjecture. Although the preceding theoretical guarantees may not hold with the same bounds because the time dependence in the upper bound is through…

We consider Bernoulli bandit models with two arms, whose means satisfy $(\mu_1,\mu_2)\in(0,1)^2$ and $\mu_1\ne\mu_2$. In the fixed-confidence setting, a sampling strategy is non-un…