arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.07995cs.LG
收藏

分支强化学习

Branching Reinforcement Learning

Tsinghua University(清华大学);Microsoft Research(微软研究院)

Yihan Du, Wei Chen

摘要:

本文提出了一种新的分支强化学习(Branching RL)模型,并研究了该模型的后悔最小化(RM)和无奖励探索(RFE)指标。与标准RL不同,其中每个事件的轨迹是单个$H$步骤路径,分支RL允许代理在状态中采取多个基本动作,使得转换相应地分支到多个后续状态,因此它生成树结构轨迹。该模型在分层推荐系统和在线广告中有重要的应用。对于分支RL,我们建立了新的Bellman方程和关键引理,即,分支值差引理和总方差的分支律,并且在指数大的轨迹下,总方差仅由$O(H^2)$约束.对于RM和RFE指标,我们分别提出了计算效率高的算法BranchVI和BranchRFE,并推导出几乎匹配的上界和下界。我们的结果是多项式的问题参数,尽管指数大的轨迹。

英文摘要:

In this paper, we propose a novel Branching Reinforcement Learning (Branching RL) model, and investigate both Regret Minimization (RM) and Reward-Free Exploration (RFE) metrics for this model. Unlike standard RL where the trajectory of each episode is a single $H$-step path, branching RL allows an agent to take multiple base actions in a state such that transitions branch out to multiple successor states correspondingly, and thus it generates a tree-structured trajectory. This model finds important applications in hierarchical recommendation systems and online advertising. For branching RL, we establish new Bellman equations and key lemmas, i.e., branching value difference lemma and branching law of total variance, and also bound the total variance by only $O(H^2)$ under an exponentially-large trajectory. For RM and RFE metrics, we propose computationally efficient algorithms BranchVI and BranchRFE, respectively, and derive nearly matching upper and lower bounds. Our results are only polynomial in problem parameters despite exponentially-large trajectories.

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。

↑