arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1303.6925math.PR
收藏

因果转移计划及其Monge-Kantorovich问题

Causal transference plans and their Monge-Kantorovich problems

Rémi Lassalle

摘要:

本文在两个具有过滤性的Polish空间的框架下研究了因果最优运输问题。具体的具体化产生的原始问题相当于随机控制和随机分析的几个经典问题;平凡的过滤产生的最优输运的常见问题。在此框架下,在标准假设下,得到了相应的变分问题的基本结果和对偶形式。这些问题本质上与鞅有关。最后,我们研究应用随机框架。具体的因果最优化问题,和随机控制的问题之间的直接等价。研究了一类随机微分方程解的最优性,证明了其唯一强解的存在性与相应的Monge问题有关.

英文摘要:

This paper investigates causal optimal transportation problems, in the framework of two Polish spaces, both endowed with filtrations. Specific concretizations yield primal problems equivalent to several classical problems of stochastic control, and of stochastic calculus ; trivial filtrations yield usual problems of optimal transport. Within this framework, primal attainments and dual formulations are obtained, under standard hypothesis, for the related variational problems. These problems are intrinsically related to martingales. Finally, we investigate applications to stochastic frameworks. A straightforward equivalence between specific causal optimization problems, and problems of stochastic control, is obtained. Solutions to a class of stochastic differential equations are characterized, as optimum to specific causal Monge-Kantorovich problems ; the existence of a unique strong solution is related to corresponding Monge 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。

↑