arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1906.09323cs.LGcs.AIcs.GTstat.ML
收藏

具有凸约束的强化学习

Reinforcement Learning with Convex Constraints

Princeton University(普林斯顿大学);University of Maryland(马里兰大学);Microsoft Research(微软研究院)

Sobhan Miryoosefi, Kianté Brantley, Hal Daumé, Miroslav Dudik, Robert Schapire

摘要:

在标准的强化学习(RL)中,学习代理试图优化整体奖励。然而,期望行为的许多关键方面更自然地表示为约束。例如,设计者可能希望限制使用不安全的动作,增加轨迹的多样性以实现探索,或者在奖励稀疏时近似专家轨迹。在本文中,我们提出了一种算法方案,可以处理RL任务中的各种约束:具体来说,任何要求某些向量测量(例如使用动作)的预期值位于凸集的约束。这捕获了以前研究过的约束(例如安全性和与专家的接近度),但也启用了新的约束类别(例如多样性)。我们的方法具有严格的理论保证,并且仅依赖于近似解决标准RL任务的能力。因此,它可以很容易地适应任何无模型或基于模型的RL。在我们的实验中,我们表明,它匹配以前的算法,通过约束执行安全,但也可以执行新的属性,这些算法不包括,如多样性。

英文摘要:

In standard reinforcement learning (RL), a learning agent seeks to optimize the overall reward. However, many key aspects of a desired behavior are more naturally expressed as constraints. For instance, the designer may want to limit the use of unsafe actions, increase the diversity of trajectories to enable exploration, or approximate expert trajectories when rewards are sparse. In this paper, we propose an algorithmic scheme that can handle a wide class of constraints in RL tasks: specifically, any constraints that require expected values of some vector measurements (such as the use of an action) to lie in a convex set. This captures previously studied constraints (such as safety and proximity to an expert), but also enables new classes of constraints (such as diversity). Our approach comes with rigorous theoretical guarantees and only relies on the ability to approximately solve standard RL tasks. As a result, it can be easily adapted to work with any model-free or model-based RL. In our experiments, we show that it matches previous algorithms that enforce safety via constraints, but can also enforce new properties that these algorithms do not incorporate, such as diversity.

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。

↑