arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2003.10526math.DGcs.ITmath-phmath.ITmath.MP
收藏

基于传输信息几何的Hessian度量

Hessian metric via transport information geometry

Wuchen Li

摘要:

本文研究了概率测度空间上赋Wasserstein $2$ -度量的泛函的Hessian度量。我们称之为传输Hessian度量,它包含并扩展了经典的Wasserstein- $2$度量.我们制定了几个动力系统与运输海森度量。发现了迁移海森度量与数学物理方程之间的若干联系。例如,在一个示例中,输运Hessian梯度流(包括Newton流)形成平均场核Stein变分梯度流,输运Boltzmann-Shannon熵的Hessian Hamilton流形成浅水方程,输运Fisher信息的Hessian梯度流形成热方程.几个例子和封闭形式的解决方案,运输海森距离。

英文摘要:

We propose to study the Hessian metric of a functional on the space of probability measures endowed with the Wasserstein $2$-metric. We name it transport Hessian metric, which contains and extends the classical Wasserstein-$2$ metric. We formulate several dynamical systems associated with transport Hessian metrics. Several connections between transport Hessian metrics and mathematical physics equations are discovered. E.g., the transport Hessian gradient flow, including Newton's flow, formulates a mean-field kernel Stein variational gradient flow; The transport Hessian Hamiltonian flow of Boltzmann-Shannon entropy forms the Shallow water equation; The transport Hessian gradient flow of Fisher information leads to the heat equation. Several examples and closed-form solutions for transport Hessian distances are presented.

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。

↑