arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.18591cs.LGcs.GTmath.STstat.TH
收藏

具有图反馈的随机上下文赌博机:从独立数到MAS数

Stochastic contextual bandits with graph feedback: from independence number to MAS number

Courant Institute of Mathematical Sciences, New York University(纽约大学柯朗数学科学研究所);Stern School of Business, New York University(纽约大学斯特恩商学院)

Yuxiao Wen, Yanjun Han, Zhengyuan Zhou

AI总结:

本文研究具有图反馈的上下文赌博机,提出图论量β_M(G)刻画学习极限,介于独立数与MAS数之间,并给出实现近乎最优遗憾的算法,表明MAS数决定统计复杂度。

摘要:

我们考虑具有图反馈的上下文赌博机,这是一类比普通上下文赌博机具有更丰富结构的交互式学习问题,其中采取一个行动会揭示反馈图中所有相邻行动在所有上下文下的奖励。与多臂赌博机设置不同,在多臂赌博机中,大量文献已对图反馈有了近乎完整的理解,而在上下文赌博机对应问题中,许多方面仍有待探索。在本文中,我们通过建立遗憾下界 $\Omega(\sqrt{\beta_M(G) T})$ 来推进这一研究,其中 $M$ 是上下文数量,$G$ 是反馈图,$\beta_M(G)$ 是我们提出的图论量,它刻画了此类问题的基本学习极限。有趣的是,$\beta_M(G)$ 在独立数 $\alpha(G)$ 和最大无环子图(MAS)数 $\mathsf{m}(G)$ 之间随上下文数量 $M$ 的变化而插值。我们还提供了在重要类别的上下文序列和/或反馈图上实现近乎最优遗憾的算法,例如在拍卖和库存控制中有应用的传递闭包图。特别是,在上下文众多的情况下,我们的结果表明,MAS数本质上刻画了上下文赌博机的统计复杂度,而非多臂赌博机中的独立数。

英文摘要:

We consider contextual bandits with graph feedback, a class of interactive learning problems with richer structures than vanilla contextual bandits, where taking an action reveals the rewards for all neighboring actions in the feedback graph under all contexts. Unlike the multi-armed bandits setting where a growing literature has painted a near-complete understanding of graph feedback, much remains unexplored in the contextual bandits counterpart. In this paper, we make inroads into this inquiry by establishing a regret lower bound $Ω(\sqrt{β_M(G) T})$, where $M$ is the number of contexts, $G$ is the feedback graph, and $β_M(G)$ is our proposed graph-theoretic quantity that characterizes the fundamental learning limit for this class of problems. Interestingly, $β_M(G)$ interpolates between $α(G)$ (the independence number of the graph) and $\mathsf{m}(G)$ (the maximum acyclic subgraph (MAS) number of the graph) as the number of contexts $M$ varies. We also provide algorithms that achieve near-optimal regret for important classes of context sequences and/or feedback graphs, such as transitively closed graphs that find applications in auctions and inventory control. In particular, with many contexts, our results show that the MAS number essentially characterizes the statistical complexity for contextual bandits, as opposed to the independence number in multi-armed bandits.

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。

↑