arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.00953stat.MLcs.LG
收藏

线性盗贼的纯探索游戏化

Gamification of Pure Exploration for Linear Bandits

INRIA - DIENS - PSL Research University(法国国家信息与自动化研究所 - 巴黎高等师范学院计算机系 - PSL研究大学);INRIA(法国国家信息与自动化研究所);DeepMind Paris(深度思维巴黎分公司)

Rémy Degenne, Pierre Ménard, Xuedong Shang, Michal Valko

摘要:

我们调查了一个积极的纯探索设置,其中包括最佳臂识别,在线性随机土匪的背景下。虽然渐近最优算法存在的标准多臂土匪,这种算法的存在最好的武器识别线性土匪一直难以捉摸,尽管几次尝试解决it. First,我们提供了一个彻底的比较和新的见解在不同的概念的最优性的线性情况下,包括G-最优性,最佳试验设计和渐近最优性的转导最优性。其次,我们设计了线性强盗中固定置信纯探索的第一个渐近最优算法。因此,我们的算法自然绕过了一个简单但困难的实例所造成的陷阱,大多数以前的算法必须被设计来显式处理。最后,我们避免了需要完全解决的最优设计问题,提供了一种方法,需要一个有效的实施。

英文摘要:

We investigate an active pure-exploration setting, that includes best-arm identification, in the context of linear stochastic bandits. While asymptotically optimal algorithms exist for standard multi-arm bandits, the existence of such algorithms for the best-arm identification in linear bandits has been elusive despite several attempts to address it. First, we provide a thorough comparison and new insight over different notions of optimality in the linear case, including G-optimality, transductive optimality from optimal experimental design and asymptotic optimality. Second, we design the first asymptotically optimal algorithm for fixed-confidence pure exploration in linear bandits. As a consequence, our algorithm naturally bypasses the pitfall caused by a simple but difficult instance, that most prior algorithms had to be engineered to deal with explicitly. Finally, we avoid the need to fully solve an optimal design problem by providing an approach that entails an efficient implementation.

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。

↑