arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.07562cs.LGstat.ML
收藏

基于无悔学习器的线性Bands最佳臂识别

Explicit Best Arm Identification in Linear Bandits Using No-Regret Learners

Indian Institute of Science(印度科学学院);Technion, Israel Institute of Technology(以色列理工学院)

Mohammadi Zaki, Avi Mohan, Aditya Gopalan

摘要:

研究了线性参数化多臂强盗的最佳臂识别问题。给定一组特征向量$\mathcal{X}\subset\mathbb{R}^d,$、置信参数$δ$和未知向量$θ^*,$,目标是识别$\arg\max_{x\in\mathcal{X}}x^Tθ^*$,概率至少为$1-δ,$,使用$x^Tθ^*.$形式的噪声测量。对于此固定置信度($δ$ -PAC)设置,我们提出了一种明确可实现且可证明的阶数最优样本复杂度算法来解决此问题。以前的方法依赖于访问极大极小优化预言机。该算法,我们称之为\textit{Phased Elimination Linear Exploration Game}(PELEG),保持一个高概率的信心椭圆包含$θ^*$在每一轮,并使用它来消除次优武器的阶段。PELEG实现了沿着最令人困惑的(即,接近,但不是最佳的)方向,通过解释问题作为两个玩家零和游戏,并使用低后悔学习器计算玩家的策略,在每一轮顺序收敛到它的鞍点。我们分析了PELEG的样本复杂性,并表明它匹配,为了,在线性强盗设置样本复杂性的实例相关的下界。我们还提供了数值计算结果,该算法符合其理论保证。

英文摘要:

We study the problem of best arm identification in linearly parameterised multi-armed bandits. Given a set of feature vectors $\mathcal{X}\subset\mathbb{R}^d,$ a confidence parameter $δ$ and an unknown vector $θ^*,$ the goal is to identify $\arg\max_{x\in\mathcal{X}}x^Tθ^*$, with probability at least $1-δ,$ using noisy measurements of the form $x^Tθ^*.$ For this fixed confidence ($δ$-PAC) setting, we propose an explicitly implementable and provably order-optimal sample-complexity algorithm to solve this problem. Previous approaches rely on access to minimax optimization oracles. The algorithm, which we call the \textit{Phased Elimination Linear Exploration Game} (PELEG), maintains a high-probability confidence ellipsoid containing $θ^*$ in each round and uses it to eliminate suboptimal arms in phases. PELEG achieves fast shrinkage of this confidence ellipsoid along the most confusing (i.e., close to, but not optimal) directions by interpreting the problem as a two player zero-sum game, and sequentially converging to its saddle point using low-regret learners to compute players' strategies in each round. We analyze the sample complexity of PELEG and show that it matches, up to order, an instance-dependent lower bound on sample complexity in the linear bandit setting. We also provide numerical results for the proposed algorithm consistent with its theoretical guarantees.

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。

↑