Spectral-scaling quasi-Newton methods with updates from the one parameter of the Broyden family

DOI10.1016/J.CAM.2013.01.012MaRDI QIDQ1946210zbMATH OpenOpenAlexFDO

Authors Zixin Chen, Wanyou Cheng

Publication date 18 April 2013

Published in Journal of Computational and Applied Mathematics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.cam.2013.01.012

zbMATH Keywords

Broyden familyglobal convergencenumerical examplesquadratic termination propertyquasi-Newton methodsingle-step convergencespectral-scalingsteepest descent methodunconstrained optimization

Mathematics Subject Classification ID

Numerical mathematical programming methods (65K05) Nonlinear programming (90C30) Methods of quasi-Newton type (90C53)

Review text

The authors consider the unconstrained optimization problem \[ \min f(x),\quad x\in\mathbb{R}^n, \] and propose spectral-scaling one parameter Broyden family methods which allow for negative values of the parameter. It is shown that the proposed methods possess some good properties such as quadratic termination property and single-step convergence rate not inferior to that of the steepest descent method when minimizing an \(n\)-dimensional quadratic function. Numerical examples are given.

Recommendations

• Quasi-Newton Algorithms with Updates from the Preconvex Part of Broyden's Family
• Memoryless quasi-Newton methods based on spectral-scaling Broyden family for unconstrained optimization
• Memoryless quasi-Newton methods based on the spectral-scaling Broyden family for Riemannian optimization
• Compact representation of the full Broyden class of quasi‐Newton updates
• Modified spectral conjugate gradient methods based on the quasi-Newton aspects
• scientific article; zbMATH DE number 2167556
• scientific article; zbMATH DE number 527494
• Global convergence of quasi-Newton methods based on adjoint Broyden updates
• A Family of Quasi-Newton Methods for Nonlinear Equations with Direct Secant Updates of Matrix Factorizations
• The convergence of quasi-Newton matrices generated by the self-scaling symmetric rank one update

Cited in

(8)

• Memoryless quasi-Newton methods based on spectral-scaling Broyden family for unconstrained optimization
• Computational experiments with scaled initial hessian approximation for the broyden family methods^∗
• scientific article; zbMATH DE number 1159073 (Why is no real title available?)
• Compact representation of the full Broyden class of quasi‐Newton updates
• Spectral scaling BFGS method
• Memoryless quasi-Newton methods based on the spectral-scaling Broyden family for Riemannian optimization
• Modified memoryless spectral-scaling Broyden family on Riemannian manifolds
• A matrix form of spectral scaling in quasi-Newton algorithm

This page was built for publication: Spectral-scaling quasi-Newton methods with updates from the one parameter of the Broyden family

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1946210)