Trust-region methods on Riemannian manifolds (Q2385533): Difference between revisions

From MaRDI portal
RedirectionBot (talk | contribs)
Changed an Item
Import240304020342 (talk | contribs)
Set profile property.
Property / MaRDI profile type
 
Property / MaRDI profile type: MaRDI publication profile / rank
 
Normal rank

Revision as of 06:57, 5 March 2024

scientific article
Language Label Description Also known as
English
Trust-region methods on Riemannian manifolds
scientific article

    Statements

    Trust-region methods on Riemannian manifolds (English)
    0 references
    0 references
    0 references
    0 references
    12 October 2007
    0 references
    This article presents an extensive study of trust-region methods on Riemannian manifolds. In the first section, an introductory overview of the background literature is presented, including existing algorithms for solving optimization problems on manifolds. The background information on Riemannian trust-region approaches in also outlined. In the second section the general framework proposed by the authors for the Riemannian trust-region algorithm is presented. This is followed by a discussion and a method for computing the trust-region step. The fourth section concentrates on issues relating to convergence analysis. The study of global and local convergence is performed in a series of theorems with proofs. In the next section an extensive list of potential applications of the proposed scheme is presented and analyzed. The article concludes with a list of useful, relevant references.
    0 references
    numerical optimization
    0 references
    trust region
    0 references
    truncated conjugate gradient
    0 references
    optimization on manifolds
    0 references
    Steihaug-Toint method
    0 references
    Riemannian manifolds
    0 references
    algorithm
    0 references
    convergence
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references
    0 references