arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1901.08708cs.LGcs.DSstat.ML
收藏

近似玻尔兹曼探索

Almost Boltzmann Exploration

University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)

Harsh Gupta, Seo Taek Kong, R. Srikant, Weina Wang

摘要:

玻尔兹曼探索被广泛用于强化学习,以提供探索和利用之间的权衡。最近,在(Cesa-Bianchi等人,2017)已经表明,从遗憾的角度来看,纯粹的玻尔兹曼探索并没有表现得很好,即使是在最简单的随机多臂强盗(MAB)问题中。在本文中,我们表明,一个简单的修改玻尔兹曼探索,激励的变化的标准加倍技巧,实现$O(K\log^{1+α} T)$遗憾的随机MAB问题的$K$武器,其中$α>0$是一个参数的算法。这改进了(Cesa-Bianchi等人,2017年),其中一个算法的灵感来自Gumbel-softmax技巧实现$O(K\log^2 T)$遗憾。我们还表明,我们的算法实现了$O(β(G) \log^{1+α} T)$遗憾的随机MAB问题与图形结构的反馈,没有知识的图形结构,其中$β(G)$是独立数的反馈图。此外,我们提出了广泛的实验结果,真实的数据集和应用程序的多臂土匪与传统的土匪反馈和图形结构的反馈。在所有情况下,我们的算法执行以及或优于国家的最先进的。

英文摘要:

Boltzmann exploration is widely used in reinforcement learning to provide a trade-off between exploration and exploitation. Recently, in (Cesa-Bianchi et al., 2017) it has been shown that pure Boltzmann exploration does not perform well from a regret perspective, even in the simplest setting of stochastic multi-armed bandit (MAB) problems. In this paper, we show that a simple modification to Boltzmann exploration, motivated by a variation of the standard doubling trick, achieves $O(K\log^{1+α} T)$ regret for a stochastic MAB problem with $K$ arms, where $α>0$ is a parameter of the algorithm. This improves on the result in (Cesa-Bianchi et al., 2017), where an algorithm inspired by the Gumbel-softmax trick achieves $O(K\log^2 T)$ regret. We also show that our algorithm achieves $O(β(G) \log^{1+α} T)$ regret in stochastic MAB problems with graph-structured feedback, without knowledge of the graph structure, where $β(G)$ is the independence number of the feedback graph. Additionally, we present extensive experimental results on real datasets and applications for multi-armed bandits with both traditional bandit feedback and graph-structured feedback. In all cases, our algorithm performs as well or better than the state-of-the-art.

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。

↑