arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1202.2277math.STmath.PRstat.TH
收藏

半有界支撑模型Bandit算法的瞬时遗憾界

Finite-time Regret Bound of a Bandit Algorithm for the Semi-bounded Support Model

Junya Honda, Akimichi Takemura

摘要:

本文考虑随机多臂强盗问题。最近提出了一种策略DMED,并证明了该策略可以使模型的每个报酬分布在一个已知的有界区间(例如[0,1])内得到支持。然而,推导出的遗憾界描述的渐近形式和有限时间内的性能一直是未知的。我们检查这一政策,并得出一个有限的时间遗憾的限制,通过精炼大偏差概率到一个简单的有限形式。此外,这一观察表明,对下界的支持的假设是不必要的,可以用一个较弱的,矩母函数的存在取代。

英文摘要:

In this paper we consider stochastic multiarmed bandit problems. Recently a policy, DMED, is proposed and proved to achieve the asymptotic bound for the model that each reward distribution is supported in a known bounded interval, e.g. [0,1]. However, the derived regret bound is described in an asymptotic form and the performance in finite time has been unknown. We inspect this policy and derive a finite-time regret bound by refining large deviation probabilities to a simple finite form. Further, this observation reveals that the assumption on the lower-boundedness of the support is not essential and can be replaced with a weaker one, the existence of the moment generating function.

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。

↑