Aldo Pacchiano, Mohammad Ghavamzadeh, Peter Bartlett, Heinrich Jiang
TL;DR本文研究了一个约束的上下文线性赌博机问题，提出了一种算法 OPLB 并证明了其 T 轮后悔度的上限，针对多臂赌博机情况提出了高效算法，同时给出了问题的下限和模拟结果。
Abstract
We study a constrained contextual linear bandit setting, where the goal of
the agent is to produce a sequence of policies, whose expected cumulative
reward over the course of $T$ rounds is maximum, and each has an expected cost
below a certain threshold $\tau$. We propose an upper-conf