Calculus of Variations and Geometric Measure Theory
home | mail | papers | authors | news | seminars | events | open positions | login

L. Ambrosio

Steepest descent flows and applications to spaces of probability measures

created by ambrosio on 12 Sep 2005


Accepted Paper

Inserted: 12 sep 2005

Year: 2005

Proceedings of the Summer School in Santander, Spain, June 2004


In these notes we summarize some of the main results of a Birkhauser book on this topic (written jointly with N.Gigli and G.Savare'), where we examine in detail the theory of curves of maximal slope in a general metric setting, following some ideas introduced by De Giorgi and collaborators in the early 80's. We study in detail the case of the Wasserstein space of probability measures. In the first part we derive new general conditions ensuring convergence of the implicit time discretization scheme to a curve of maximal slope, the uniqueness, and the error estimates.

In the second part we study in detail the differentiable structure of the Wasserstein space, to which the metric theory applies, and use this structure to give also an equivalent concept of gradient flow. Our analysis includes measures in infinite-dimensional Hilbert spaces and it does not require any absolute continuity assumption on the measures involved.

Keywords: Wasserstein distance, Gradient flows, Steepest descent flows


Credits | Cookie policy | HTML 5 | CSS 2.1