arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2103.05750cs.LGstat.ML
收藏

参数漂移下广义线性Bands的后悔界

Regret Bounds for Generalized Linear Bandits under Parameter Drift

LTCI;TélécomParis(巴黎电信学院);ENS Paris(巴黎高等师范学院);Université PSL(巴黎文理大学);CNRS(法国国家科学研究中心);Inria(法国国家信息与自动化研究所);Criteo AI Lab(Criteo人工智能实验室);Criteo(Criteo公司)

Louis Faury, Yoan Russac, Marc Abeille, Clément Calauzènes

摘要:

广义线性强盗(GLB)是线性强盗(LB)设置的强大扩展,扩大了奖励参数化的好处。在本文中,我们研究GLB在非平稳环境中,其特征在于一个通用的度量的非平稳性称为变化预算或\emph{parameter-drift},表示$B_T$。虽然以前的尝试已经作出了扩展LB算法,以这种设置,他们忽略了GLB的一个显着特点,缺陷他们的结果。在这项工作中,我们引入了一个新的算法,解决了这个困难。我们证明了在一个几何假设下的行动集,我们的方法享有$\tilde{\mathcal{O}}(B_T^{1/3}T^{2/3})$遗憾界。在一般情况下,我们表明,它遭受最多的$\tilde{\mathcal{O}}(B_T^{1/5}T^{4/5})$遗憾。我们贡献的核心是Filippi等人(2010)中引入的投影步骤的推广,适用于问题的非平稳性质。我们的分析揭示了从设置继承的中央机制,明确分裂的学习和跟踪方面的问题的治疗。

英文摘要:

Generalized Linear Bandits (GLBs) are powerful extensions to the Linear Bandit (LB) setting, broadening the benefits of reward parametrization beyond linearity. In this paper we study GLBs in non-stationary environments, characterized by a general metric of non-stationarity known as the variation-budget or \emph{parameter-drift}, denoted $B_T$. While previous attempts have been made to extend LB algorithms to this setting, they overlook a salient feature of GLBs which flaws their results. In this work, we introduce a new algorithm that addresses this difficulty. We prove that under a geometric assumption on the action set, our approach enjoys a $\tilde{\mathcal{O}}(B_T^{1/3}T^{2/3})$ regret bound. In the general case, we show that it suffers at most a $\tilde{\mathcal{O}}(B_T^{1/5}T^{4/5})$ regret. At the core of our contribution is a generalization of the projection step introduced in Filippi et al. (2010), adapted to the non-stationary nature of the problem. Our analysis sheds light on central mechanisms inherited from the setting by explicitly splitting the treatment of the learning and tracking aspects of the problem.

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。

↑