arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1608.03023cs.LGstat.ML
收藏

随机一阶盗贼

Stochastic Rank-1 Bandits

University of Wisconsin-Madison(威斯康星大学麦迪逊分校);Adobe Research(Adobe研究院);University of Alberta(阿尔伯塔大学);Telecom ParisTech(巴黎高等电信学院)

Sumeet Katariya, Branislav Kveton, Csaba Szepesvari, Claire Vernade, Zheng Wen

摘要:

我们提出了随机秩- $1$强盗,一类在线学习问题,在每一步的学习代理选择一对行和列武器,并收到他们的价值作为奖励的产品。这个问题的主要挑战是行和列的各个值是不可观察的。我们假设这些值是随机的,并且是独立绘制的。我们提出了一个计算效率高的算法来解决我们的问题,我们称之为Rank 1 Elim。我们推导出一个$O((K + L) (1 / Δ) \log n)$上界的$n$ -步遗憾,其中$K$是行数,$L$是列数,和$Δ$是最小的行和列的差距,在假设的平均行和列奖励有界远离零。据我们所知,我们提出了第一个强盗算法,找到一个秩$1$矩阵的最大条目,其遗憾是线性的$K + L$,$1 / Δ$,和$\log n$。我们还得到了一个几乎匹配的下界。最后,我们评估Rank 1 Elim经验上的多个问题。我们观察到,它利用了我们问题的结构,即使我们的建模假设被轻微违反,它也可以学习接近最优的解决方案。

英文摘要:

We propose stochastic rank-$1$ bandits, a class of online learning problems where at each step a learning agent chooses a pair of row and column arms, and receives the product of their values as a reward. The main challenge of the problem is that the individual values of the row and column are unobserved. We assume that these values are stochastic and drawn independently. We propose a computationally-efficient algorithm for solving our problem, which we call Rank1Elim. We derive a $O((K + L) (1 / Δ) \log n)$ upper bound on its $n$-step regret, where $K$ is the number of rows, $L$ is the number of columns, and $Δ$ is the minimum of the row and column gaps; under the assumption that the mean row and column rewards are bounded away from zero. To the best of our knowledge, we present the first bandit algorithm that finds the maximum entry of a rank-$1$ matrix whose regret is linear in $K + L$, $1 / Δ$, and $\log n$. We also derive a nearly matching lower bound. Finally, we evaluate Rank1Elim empirically on multiple problems. We observe that it leverages the structure of our problems and can learn near-optimal solutions even if our modeling assumptions are mildly violated.

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。

↑