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 2010.01014cs.LGcs.SYeess.SYmath.OCstat.ML
收藏

连续时间和离散空间中的POMDPs

POMDPs in Continuous Time and Discrete Spaces

Technische Universität Darmstadt(达姆施塔特工业大学)

Bastian Alt, Matthias Schultheis, Heinz Koeppl

摘要:

许多过程,如工程中的离散事件系统或生物学中的种群动力学,在离散空间和连续时间中进化。我们考虑的问题,最优决策的离散状态和行动空间系统的部分可观性。这使我们的工作处于最优滤波和最优控制的交叉点。在目前的研究状态下,仍然缺乏一个数学描述的同时决策和过滤在连续时间有限的状态和行动空间。本文给出了连续时间部分可观测马尔可夫决策过程(POMDP)的数学描述。通过利用最优滤波理论,我们推导出一个哈密尔顿-雅可比-贝尔曼(HJB)型方程,其特征在于最优解。使用来自深度学习的技术,我们近似地求解所得到的偏积分微分方程。我们提出了(i)通过学习价值函数的近似值离线解决决策问题的方法和(ii)使用深度强化学习在置信空间中提供解决方案的在线算法。我们展示了一组玩具的例子,为未来的方法提供高维问题的解决方案铺平了道路的适用性。

英文摘要:

Many processes, such as discrete event systems in engineering or population dynamics in biology, evolve in discrete space and continuous time. We consider the problem of optimal decision making in such discrete state and action space systems under partial observability. This places our work at the intersection of optimal filtering and optimal control. At the current state of research, a mathematical description for simultaneous decision making and filtering in continuous time with finite state and action spaces is still missing. In this paper, we give a mathematical description of a continuous-time partial observable Markov decision process (POMDP). By leveraging optimal filtering theory we derive a Hamilton-Jacobi-Bellman (HJB) type equation that characterizes the optimal solution. Using techniques from deep learning we approximately solve the resulting partial integro-differential equation. We present (i) an approach solving the decision problem offline by learning an approximation of the value function and (ii) an online algorithm which provides a solution in belief space using deep reinforcement learning. We show the applicability on a set of toy examples which pave the way for future methods providing solutions for high dimensional problems.

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。

↑