The Intrinsic Riemannian Proximal Gradient Method for Nonconvex Optimization

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Bergmann, Ronny, Jasa, Hajg, John, Paula, Pfeffer, Max
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916790689333248
author Bergmann, Ronny
Jasa, Hajg
John, Paula
Pfeffer, Max
author_facet Bergmann, Ronny
Jasa, Hajg
John, Paula
Pfeffer, Max
contents We consider the proximal gradient method on Riemannian manifolds for functions that are possibly not geodesically convex. Starting from the forward-backward-splitting, we define an intrinsic variant of the proximal gradient method that uses proximal maps defined on the manifold and therefore does not require or work in the embedding. We investigate its convergence properties and illustrate its numerical performance, particularly for nonconvex or nonembedded problems that are hence out of reach for other methods.
format Preprint
id arxiv_https___arxiv_org_abs_2506_09775
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle The Intrinsic Riemannian Proximal Gradient Method for Nonconvex Optimization
Bergmann, Ronny
Jasa, Hajg
John, Paula
Pfeffer, Max
Optimization and Control
Numerical Analysis
Differential Geometry
90C26, 49Q99, 49M30, 65K10
We consider the proximal gradient method on Riemannian manifolds for functions that are possibly not geodesically convex. Starting from the forward-backward-splitting, we define an intrinsic variant of the proximal gradient method that uses proximal maps defined on the manifold and therefore does not require or work in the embedding. We investigate its convergence properties and illustrate its numerical performance, particularly for nonconvex or nonembedded problems that are hence out of reach for other methods.
title The Intrinsic Riemannian Proximal Gradient Method for Nonconvex Optimization
topic Optimization and Control
Numerical Analysis
Differential Geometry
90C26, 49Q99, 49M30, 65K10
url https://arxiv.org/abs/2506.09775