arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2011.02565cs.LGcs.AI
收藏

丰富多样性的选项评论家

Diversity-Enriched Option-Critic

Anand Kamat, Doina Precup

AI总结:

研究针对选项评论家框架存在的问题,提出通过学习多样选项解决。引入信息论内在奖励和新终止目标,在离散和连续控制任务上实现端到端学习选项,相比选项评论家大幅提升,还能生成强大、可复用、可靠且可解释的选项。

摘要:

时间抽象使强化学习智能体能够在不同时间尺度上表示知识并制定策略。选项评论家框架已被证明能在无模型设置中以端到端方式学习表示为选项的时间扩展动作。然而,由于两个主要挑战,选项评论家的可行性仍然有限,即多个选项行为非常相似,或与任务相关的选项集不断缩小。这些情况不仅消除了时间抽象的必要性,还影响性能。本文通过学习一组多样的选项来解决这些问题。我们引入了一种信息论内在奖励来增强任务奖励,以及一个新的终止目标,以鼓励选项集中的行为多样性。我们通过实验表明,我们提出的方法能够在几个离散和连续控制任务上以端到端方式学习选项,比选项评论家有大幅提升。此外,与选项评论家相比,我们的方法可持续地生成强大、可复用、可靠且可解释的选项。

英文摘要:

Temporal abstraction allows reinforcement learning agents to represent knowledge and develop strategies over different temporal scales. The option-critic framework has been demonstrated to learn temporally extended actions, represented as options, end-to-end in a model-free setting. However, feasibility of option-critic remains limited due to two major challenges, multiple options adopting very similar behavior, or a shrinking set of task relevant options. These occurrences not only void the need for temporal abstraction, they also affect performance. In this paper, we tackle these problems by learning a diverse set of options. We introduce an information-theoretic intrinsic reward, which augments the task reward, as well as a novel termination objective, in order to encourage behavioral diversity in the option set. We show empirically that our proposed method is capable of learning options end-to-end on several discrete and continuous control tasks, outperforms option-critic by a wide margin. Furthermore, we show that our approach sustainably generates robust, reusable, reliable and interpretable options, in contrast to option-critic.

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。

↑