论文标题
重新审视的前两种算法
Top Two Algorithms Revisited
论文作者
论文摘要
出现了前两种算法，作为汤普森采样对多臂匪徒模型中最佳手臂识别的适应（Russo，2016），用于武器的参数家族。他们通过在两个候选臂，一个领导者和一个挑战者中随机化来选择要采样的下一个手臂。尽管具有良好的经验表现，但仅当手臂是具有已知方差的高斯时，才能获得固定信心最佳臂识别的理论保证。在本文中，我们提供了前两种方法的一般分析，该方法确定了领导者，挑战者和武器（可能是非参数）分布的理想特性。结果，我们获得了理论上支持的前两种算法，用于具有有限分布的最佳臂识别。我们的证明方法特别证明了用于选择从汤普森采样继承的领导者的采样步骤可以用其他选择代替，例如选择经验最佳臂。
Top Two algorithms arose as an adaptation of Thompson sampling to best arm identification in multi-armed bandit models (Russo, 2016), for parametric families of arms. They select the next arm to sample from by randomizing among two candidate arms, a leader and a challenger. Despite their good empirical performance, theoretical guarantees for fixed-confidence best arm identification have only been obtained when the arms are Gaussian with known variances. In this paper, we provide a general analysis of Top Two methods, which identifies desirable properties of the leader, the challenger, and the (possibly non-parametric) distributions of the arms. As a result, we obtain theoretically supported Top Two algorithms for best arm identification with bounded distributions. Our proof method demonstrates in particular that the sampling step used to select the leader inherited from Thompson sampling can be replaced by other choices, like selecting the empirical best arm.
