Extended Levenberg-Marquardt Method for Composite Function Minimization

Extended Levenberg-Marquardt Method for Composite Function Minimization

Year:    2017

Author:    Jianchao Huang, Zaiwen Wen, Xiantao Xiao

Journal of Computational Mathematics, Vol. 35 (2017), Iss. 4 : pp. 529–546

Abstract

In this paper, we propose an extended Levenberg-Marquardt (ELM) framework that generalizes the classic Levenberg-Marquardt (LM) method to solve the unconstrained minimization problem min $ρ(r(x))$, where $r$ : $\mathbb{R}^n$ → $\mathbb{R}^m$ and $ρ$ : $\mathbb{R}^m$ → $\mathbb{R}$. We also develop a few inexact variants which generalize ELM to the cases where the inner subproblem is not solved exactly and the Jacobian is simplified, or perturbed. Global convergence and local superlinear convergence are established under certain suitable conditions. Numerical results show that our methods are promising.

You do not have full access to this article.

Already a Subscriber? Sign in as an individual or via your institution

Journal Article Details

Publisher Name:    Global Science Press

Language:    English

DOI:    https://doi.org/10.4208/jcm.1702-m2016-0699

Journal of Computational Mathematics, Vol. 35 (2017), Iss. 4 : pp. 529–546

Published online:    2017-01

AMS Subject Headings:   

Copyright:    COPYRIGHT: © Global Science Press

Pages:    18

Keywords:    Unconstrained minimization Composite function Levenberg-Marquardt method.

Author Details

Jianchao Huang

Zaiwen Wen

Xiantao Xiao

  1. Convergence analysis of a subsampled Levenberg-Marquardt algorithm

    Xing, Ganchen | Gu, Jian | Xiao, Xiantao

    Operations Research Letters, Vol. 51 (2023), Iss. 4 P.379

    https://doi.org/10.1016/j.orl.2023.05.005 [Citations: 3]
  2. Efficient ptychographic phase retrieval via a matrix-free Levenberg-Marquardt algorithm

    Kandel, Saugat | Maddali, S. | Nashed, Youssef S. G. | Hruszkewycz, Stephan O. | Jacobsen, Chris | Allain, Marc

    Optics Express, Vol. 29 (2021), Iss. 15 P.23019

    https://doi.org/10.1364/OE.422768 [Citations: 7]