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 2412.03478math.OCcs.NAmath.NA
收藏

通过概率测度的希尔伯特空间嵌入求解蒙日问题

Solving Monge problem by Hilbert space embeddings of probability measures

Department of Mathematical and Computing Science, Institute of Science Tokyo(东京科学大学数学与计算科学系)

Takafumi Saito, Yumiharu Nakano

AI总结:

本文提出基于希尔伯特空间嵌入的概率测度深度学习方法,用于求解大规模最优质量运输问题,并证明其在$ L^2 $成本下的收敛性。

摘要:

我们提出了一种深度学习方法来解决经典的蒙日最优质量运输问题,其中分布约束被处理为由概率测度的希尔伯特空间嵌入理论中的最大均值差异定义的惩罚项。我们证明了所提出方法生成的传输映射在$ L^2 $成本问题中收敛到最优传输映射。几个数值实验验证了我们的方法。特别是,我们展示了我们的方法可以应用于大规模蒙日问题。这是ICORES 2025会议论文的更正版本。

英文摘要:

We propose deep learning methods for classical Monge's optimal mass transportation problems, where where the distribution constraint is treated as penalty terms defined by the maximum mean discrepancy in the theory of Hilbert space embeddings of probability measures. We prove that the transport maps given by the proposed methods converge to optimal transport maps in the problem with $L^2$ cost. Several numerical experiments validate our methods. In particular, we show that our methods are applicable to large-scale Monge problems. This is a corrected version of the ICORES 2025 proceedings paper.

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。

↑