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 2211.07488math.OC
收藏

参数转移分布下的鲁棒马尔可夫决策过程

Robust Markov decision processes under parametric transition distributions

Ben Black, Trivikram Dokka, Christopher Kirkbride

摘要:

研究了参数转移分布下的鲁棒马尔可夫决策过程。我们假设真实的过渡分布是由某些参数分布唯一指定的,并明确要求模型的最差情况分布是由同一参数族中的分布唯一指定的。在制定参数鲁棒模型,我们专注于开发算法进行强大的贝尔曼更新完成强大的价值迭代。我们首先制定的更新作为一个线性规划离散的模糊集。由于这个模型的规模与问题的大小,需要大量的预计算差,我们开发了两个额外的算法来解决强大的贝尔曼更新。首先,我们提出了一个切割面算法,在较短的时间内解决这个线性规划。该算法需要相同的预计算,但仅在模糊度集的小子集上求解线性规划。其次,我们提出了一种新的基于投影的二分法搜索算法,完全消除了离散化的需要,不需要任何预先计算。我们在二项和泊松需求下的动态多周期报童问题上广泛测试了我们的算法。此外,我们比较了我们的方法与非参数的φ-散度为基础的方法从文献中。我们表明,我们的基于投影的算法完成强大的值迭代显着快于我们的其他两个参数算法,也快于其非参数等价。

英文摘要:

This paper considers robust Markov decision processes under parametric transition distributions. We assume that the true transition distribution is uniquely specified by some parametric distribution, and explicitly enforce that the worst-case distribution from the model is uniquely specified by a distribution in the same parametric family. After formulating the parametric robust model, we focus on developing algorithms for carrying out the robust Bellman updates required to complete robust value iteration. We first formulate the update as a linear program by discretising the ambiguity set. Since this model scales poorly with problem size and requires large amounts of pre-computation, we develop two additional algorithms for solving the robust Bellman update. Firstly, we present a cutting surface algorithm for solving this linear program in a shorter time. This algorithm requires the same pre-computation, but only ever solves the linear program over small subsets of the ambiguity set. Secondly, we present a novel projection-based bisection search algorithm that completely eliminates the need for discretisation and does not require any pre-computation. We test our algorithms extensively on a dynamic multi-period newsvendor problem under binomial and Poisson demands. In addition, we compare our methods with the non-parametric phi-divergence based methods from the literature. We show that our projection-based algorithm completes robust value iteration significantly faster than our other two parametric algorithms, and also faster than its non-parametric equivalent.

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。

↑