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 2404.04499math.PR
收藏

关于 $\mathbb N$ 上概率测度的基于 Fourier 的距离与 Wasserstein 距离之间的等价性

On the equivalence between Fourier-based and Wasserstein distances for probability measures on $\mathbb N$

Fei Cao, Xiaoqian Gong

AI总结:

本文证明非负整数状态空间上基于 Fourier 的离散度量与 Wasserstein 距离之间的多种等价关系,并探讨其在经济物理学模型中的应用。

摘要:

在本手稿中,我们研究离散状态空间上基于 Fourier 的度量与著名的 Wasserstein 距离之间的等价性。自从 Giuseppe Toscani 及其同事 [9, 14, 16] 在研究动力学类型偏微分方程时引入基于 Fourier 的度量以来,其在连续状态空间中的应用已十分普遍;然而,其离散类似物的引入是较近期的事情 [2],且相关研究似乎远不充分。在本工作中,我们证明了当状态空间为非负整数集合 $\mathbb N$ 时,基于 Fourier 的度量与 Wasserstein 距离之间存在多种关系。最后,我们还描述了这种度量等价性在经济物理学模型中的潜在应用,这些模型正是推动本工作的动机。

英文摘要:

In this manuscript we investigate the equivalence of Fourier-based metrics on discrete state spaces with the well-known Wasserstein distances. While the use of Fourier-based metrics in continuous state spaces is ubiquitous since its introduction by Giuseppe Toscani and his colleagues [9, 14, 16] in the study of kinetic-type partial differential equations, the introduction of its discrete analog is recent [2] and seems to be far less studied. In this work, various relations between Fourier-based metrics and Wasserstein distances are shown to hold when the state space is the set of non-negative integers $\mathbb N$. Lastly, we also describe potential applications of such equivalence of metrics in models from econophysics which motivate the present work.

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。

↑