arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1503.08370cs.LG
收藏

全球强盗

Global Bandits

University of California, Los Angeles(加利福尼亚大学洛杉矶分校);Bilkent University(比尔肯特大学)

Onur Atan, Cem Tekin, Mihaela van der Schaar

摘要:

多武装土匪(MAB)模型的顺序决策问题,其中学习者顺序选择武器未知的奖励分布,以最大限度地提高其累积奖励。大多数关于MAB的先前工作假设每个臂的奖励分布是独立的。但在各种各样的决策问题中--从药物剂量到动态定价--不同手臂的预期回报是相关的,因此选择一个手臂也提供了其他手臂预期回报的信息。我们提出并分析了一类决策问题的模型,我们称之为{ \em全球土匪}。在所有武器的奖励是一个单一的未知参数的确定性函数的情况下,我们构建了一个贪婪的政策,实现{ \em有界遗憾},与一个边界,取决于该问题的单个真参数。因此,该策略仅以概率1选择次优武器200次。对于这种情况下,我们也得到了一个约束的遗憾,是{ \em独立的真实参数};这个界是次线性的,指数取决于信息的武器。我们还提出了一个贪婪的政策,实现$\tilde{\mathcal{O}}(\sqrt{T})$最坏情况下和$\mathcal{O}(1)$参数相关的遗憾的变体。最后,我们进行实验动态定价,并表明所提出的算法实现了显着的收益相对于著名的基准。

英文摘要:

Multi-armed bandits (MAB) model sequential decision making problems, in which a learner sequentially chooses arms with unknown reward distributions in order to maximize its cumulative reward. Most of the prior work on MAB assumes that the reward distributions of each arm are independent. But in a wide variety of decision problems -- from drug dosage to dynamic pricing -- the expected rewards of different arms are correlated, so that selecting one arm provides information about the expected rewards of other arms as well. We propose and analyze a class of models of such decision problems, which we call {\em global bandits}. In the case in which rewards of all arms are deterministic functions of a single unknown parameter, we construct a greedy policy that achieves {\em bounded regret}, with a bound that depends on the single true parameter of the problem. Hence, this policy selects suboptimal arms only finitely many times with probability one. For this case we also obtain a bound on regret that is {\em independent of the true parameter}; this bound is sub-linear, with an exponent that depends on the informativeness of the arms. We also propose a variant of the greedy policy that achieves $\tilde{\mathcal{O}}(\sqrt{T})$ worst-case and $\mathcal{O}(1)$ parameter dependent regret. Finally, we perform experiments on dynamic pricing and show that the proposed algorithms achieve significant gains with respect to the well-known benchmarks.

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。

↑