arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.03039cs.LG
收藏

多面上下文Bandits:一个神经网络视角

Multi-facet Contextual Bandits: A Neural Network Perspective

University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校);Mayo Clinic Arizona(亚利桑那州梅奥诊所)

Yikun Ban, Jingrui He, Curtiss B. Cook

摘要:

上下文多臂强盗已被证明是一个有效的工具,在推荐系统。在本文中,我们研究了一个新的问题,多方面的土匪涉及一组土匪,每个用户的需求从一个独特的方面。在每一轮中,对于给定的用户,我们需要从每个强盗中选择一只手臂,这样所有手臂的组合才能最大化最终的奖励。这个问题可以立即应用于电子商务,医疗保健等,为了解决这个问题,我们提出了一种新的算法,名为MuFasa,它利用组装神经网络,共同学习多个土匪的底层奖励功能。它估计了一个置信上限(UCB)与预期的回报,以平衡开发和勘探。在适度的假设下,我们给出了MuFasa的后悔分析。它可以达到接近最优的$\widetilde{ \mathcal{O}}((K+1)\sqrt{T})$后悔界限,其中$K$是强盗的数量,$T$是玩过的回合数。此外,我们进行了大量的实验,以表明MuFasa在真实世界的数据集上优于强基线。

英文摘要:

Contextual multi-armed bandit has shown to be an effective tool in recommender systems. In this paper, we study a novel problem of multi-facet bandits involving a group of bandits, each characterizing the users' needs from one unique aspect. In each round, for the given user, we need to select one arm from each bandit, such that the combination of all arms maximizes the final reward. This problem can find immediate applications in E-commerce, healthcare, etc. To address this problem, we propose a novel algorithm, named MuFasa, which utilizes an assembled neural network to jointly learn the underlying reward functions of multiple bandits. It estimates an Upper Confidence Bound (UCB) linked with the expected reward to balance between exploitation and exploration. Under mild assumptions, we provide the regret analysis of MuFasa. It can achieve the near-optimal $\widetilde{ \mathcal{O}}((K+1)\sqrt{T})$ regret bound where $K$ is the number of bandits and $T$ is the number of played rounds. Furthermore, we conduct extensive experiments to show that MuFasa outperforms strong baselines on real-world data sets.

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。

↑