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 2204.04773stat.MLcs.LG
收藏

不完全情境下贪婪策略在强盗中的最坏情况性能

Worst-case Performance of Greedy Policies in Bandits with Imperfect Context Observations

Hongju Park, Mohamad Kazem Shirani Faradonbeh

摘要:

上下文强盗是在具有时变成分的环境中的不确定性下的序列决策的规范模型。在这种情况下,每个强盗手臂的预期奖励由一个未知参数与该手臂的上下文向量的内积组成。经典的强盗设置在很大程度上依赖于假设的上下文是完全观察到的,而不完全观察上下文强盗的更丰富的模型的研究是不成熟的。这项工作考虑了贪婪强化学习策略,这些策略采取行动,就好像参数和未观察到的上下文的当前估计值与相应的真值一致。我们建立了非渐近的最坏情况下的遗憾增长的时间范围和故障概率的多项式,而它的规模与武器的数量线性。数值分析显示上述效率的贪婪政策也提供。

英文摘要:

Contextual bandits are canonical models for sequential decision-making under uncertainty in environments with time-varying components. In this setting, the expected reward of each bandit arm consists of the inner product of an unknown parameter with the context vector of that arm. The classical bandit settings heavily rely on assuming that the contexts are fully observed, while study of the richer model of imperfectly observed contextual bandits is immature. This work considers Greedy reinforcement learning policies that take actions as if the current estimates of the parameter and of the unobserved contexts coincide with the corresponding true values. We establish that the non-asymptotic worst-case regret grows poly-logarithmically with the time horizon and the failure probability, while it scales linearly with the number of arms. Numerical analysis showcasing the above efficiency of Greedy policies is also provided.

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。

↑