The back-and-forth method for Wasserstein gradient flows

The back-and-forth method for Wasserstein gradient flows
复制标题

Wasserstein 梯度流的来回方法

DOI:
--
复制
发表时间:
2020
期刊:
E S A I M: Control, Optimisation and Calculus of Variations
影响因子:
--
通讯作者:
F. L'eger
F. L'eger
中科院分区:
--
文献类型:
--
作者:
M. Jacobs;Wonjun Lee;F. L'eger

文献摘要

参考文献

被引文献

相似文献

我们提出了一种方法来有效地计算Wasserstein梯度流。我们的方法是基于一个推广的来回的方法(BFM)介绍了雅各布斯和Léger [Numer. 146(2020)513-544.]。来解决最优运输问题。通过求解JKO格式的对偶问题,我们发展了梯度流。一般来说,对偶问题比原始问题表现得更好。这使我们能够有效地运行大规模的梯度流模拟的一大类内部能量,包括奇异和非凸能量。
We present a method to efficiently compute Wasserstein gradient flows. Our approach is based on a generalization of the back-and-forth method (BFM) introduced in Jacobs and Léger [Numer. Math. 146 (2020) 513–544.]. to solve optimal transport problems. We evolve the gradient flow by solving the dual problem to the JKO scheme. In general, the dual problem is much better behaved than the primal problem. This allows us to efficiently run large scale gradient flows simulations for a large class of internal energies including singular and non-convex energies.
最佳运输中的 $L^1$-收缩原理
DOI: 10.2422/2036-2145.202010_013
发表时间: 2022
期刊: ANNALI SCUOLA NORMALE SUPERIORE - CLASSE DI SCIENZE
影响因子: --
作者:
Jacobs, Matt;Kim, Inwon Christina;Tong, Jiajun
通讯作者: Tong, Jiajun