Solving the quadratic trust-region subproblem in a low-memory BFGS framework

DOI10.1080/10556780802413579MaRDI QIDQ3605193zbMATH OpenOpenAlexFDO

Authors

Publication date 23 February 2009

Published in Optimization Methods \& Software (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1080/10556780802413579

zbMATH Keywords

eigenvaluesL-BFGS methodlarge scale optimizationnearly exact methodnegative curvature directiontrust region

Mathematics Subject Classification ID

Direct numerical methods for linear systems and matrix inversion (65F05) Numerical computation of eigenvalues and eigenvectors of matrices (65F15) Large-scale problems in mathematical programming (90C06) Nonlinear programming (90C30)

Recommendations

• A new matrix-free algorithm for the large-scale trust-region subproblem
• A practical method for solving large-scale TRS
• On solving L-SR1 trust-region subproblems
• A curvilinear method based on minimal-memory BFGS updates
• Solving the Trust-Region Subproblem using the Lanczos Method

Cited in

(7)

• A new matrix-free algorithm for the large-scale trust-region subproblem
• On efficiently computing the eigenvalues of limited-memory quasi-Newton matrices
• An affine scaling derivative-free trust region method with interior backtracking technique for bounded-constrained nonlinear programming
• On solving L-SR1 trust-region subproblems
• Algorithm 943: MSS: MATLAB software for L-BFGS trust-region subproblems for large-scale optimization
• A practical method for solving large-scale TRS
• A modified nearly exact method for solving low-rank trust region subproblem

Describes a project that uses

Uses Software

• L-BFGS
• LSTRS

This page was built for publication: Solving the quadratic trust-region subproblem in a low-memory BFGS framework

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3605193)