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 2102.04323cs.AIcs.LG
收藏

发现一组最坏情况下的奖励策略

Discovering a set of policies for the worst case reward

DeepMind

Tom Zahavy, Andre Barreto, Daniel J Mankowitz, Shaobo Hou, Brendan O'Donoghue, Iurii Kemaev, Satinder Singh

摘要:

我们研究如何构建一组可以组合在一起的策略来解决一组强化学习任务的问题。每个任务都是一个不同的奖励函数,定义为已知特征的线性组合。我们考虑一类特定的政策组合,我们称之为集改进政策(SIP):给定一组政策和一组任务,SIP是前者的任何组合,其性能至少与其成分在所有任务中一样好。我们专注于SIP的最保守的实例化,设置最大策略(SMP),所以我们的分析扩展到任何SIP。这包括已知的策略组合操作符,如广义策略改进。我们的主要贡献是一个政策迭代算法,建立了一套政策,以最大限度地提高最坏情况下的性能所产生的SMP的一组任务。该算法的工作原理是连续添加新的策略集。我们表明,所得到的SMP严格提高在每次迭代的最坏情况下的性能,该算法只停止时,不存在一个政策,导致性能提高。我们在网格世界和DeepMind控制套件的一组域上对我们的算法进行了经验评估。我们证实了我们的理论结果单调提高我们的算法的性能。有趣的是,我们还表明,经验的算法计算的政策集是不同的,导致不同的轨迹在网格世界和非常独特的运动技能的控制套件。

英文摘要:

We study the problem of how to construct a set of policies that can be composed together to solve a collection of reinforcement learning tasks. Each task is a different reward function defined as a linear combination of known features. We consider a specific class of policy compositions which we call set improving policies (SIPs): given a set of policies and a set of tasks, a SIP is any composition of the former whose performance is at least as good as that of its constituents across all the tasks. We focus on the most conservative instantiation of SIPs, set-max policies (SMPs), so our analysis extends to any SIP. This includes known policy-composition operators like generalized policy improvement. Our main contribution is a policy iteration algorithm that builds a set of policies in order to maximize the worst-case performance of the resulting SMP on the set of tasks. The algorithm works by successively adding new policies to the set. We show that the worst-case performance of the resulting SMP strictly improves at each iteration, and the algorithm only stops when there does not exist a policy that leads to improved performance. We empirically evaluate our algorithm on a grid world and also on a set of domains from the DeepMind control suite. We confirm our theoretical results regarding the monotonically improving performance of our algorithm. Interestingly, we also show empirically that the sets of policies computed by the algorithm are diverse, leading to different trajectories in the grid world and very distinct locomotion skills in the control suite.

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。

↑