arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2310.00539stat.MLcs.LG
收藏

最佳臂识别中基于最佳挑战者规则的Thompson探索

Thompson Exploration with Best Challenger Rule in Best Arm Identification

The University of Tokyo(东京大学);RIKEN AIP(理化学研究所AIP);Kyoto University(京都大学)

Jongyeong Lee, Junya Honda, Masashi Sugiyama

AI总结:

针对固定置信度最佳臂识别问题,提出结合Thompson sampling与best challenger rule的新策略,无需强制探索且计算高效,在多臂问题上实现近最优并有竞争力的样本复杂度。

摘要:

本文研究了规范单参数指数模型中老虎机框架下的固定置信度最佳臂识别(BAI)问题。针对此问题,已提出许多策略,但大多数策略需要在每一轮求解一个优化问题,和/或被迫至少探索某个臂一定次数,除了那些仅限于高斯模型的策略。为了解决这些局限性,我们提出了一种新策略,将Thompson sampling与一种被称为best challenger rule的计算高效方法相结合。虽然Thompson sampling最初被考虑用于最大化累积奖励,但我们证明它可以自然地用于在BAI中探索臂而无需强制。我们证明,对于任意双臂老虎机问题,我们的策略是渐近最优的,并且对于$K\geq 3$的一般$K$臂老虎机问题实现了近最优。然而,在数值实验中,我们的策略在样本复杂度方面与渐近最优策略相比表现出有竞争力的性能,同时需要更少的计算成本。此外,我们通过将我们的策略与$β$-optimality概念进行比较来突出其优势,$β$-optimality是一种渐近最优性的放宽概念,通常在包括所提策略在内的一类策略的分析中被考虑。

英文摘要:

This paper studies the fixed-confidence best arm identification (BAI) problem in the bandit framework in the canonical single-parameter exponential models. For this problem, many policies have been proposed, but most of them require solving an optimization problem at every round and/or are forced to explore an arm at least a certain number of times except those restricted to the Gaussian model. To address these limitations, we propose a novel policy that combines Thompson sampling with a computationally efficient approach known as the best challenger rule. While Thompson sampling was originally considered for maximizing the cumulative reward, we demonstrate that it can be used to naturally explore arms in BAI without forcing it. We show that our policy is asymptotically optimal for any two-armed bandit problems and achieves near optimality for general $K$-armed bandit problems for $K\geq 3$. Nevertheless, in numerical experiments, our policy shows competitive performance compared to asymptotically optimal policies in terms of sample complexity while requiring less computation cost. In addition, we highlight the advantages of our policy by comparing it to the concept of $β$-optimality, a relaxed notion of asymptotic optimality commonly considered in the analysis of a class of policies including the proposed one.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑