A Class of Diagonally Preconditioned Limited Memory Quasi-Newton Methods for Large-Scale Unconstrained Optimization

The focus of this thesis is to diagonally precondition on the limited memory quasi-Newton method for large scale unconstrained optimization problem. Particularly, the centre of discussion is on diagonally preconditioned limited memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) method. L-BFGS method h...

Full description

Bibliographic Details
Main Author: Chen, Chuei Yee
Format: Thesis
Language:English
English
Published: 2009
Subjects:
Online Access:http://psasir.upm.edu.my/id/eprint/7547/
http://psasir.upm.edu.my/id/eprint/7547/1/ABS_---__FS_2009_29.pdf
_version_ 1848840625192960000
author Chen, Chuei Yee
author_facet Chen, Chuei Yee
author_sort Chen, Chuei Yee
building UPM Institutional Repository
collection Online Access
description The focus of this thesis is to diagonally precondition on the limited memory quasi-Newton method for large scale unconstrained optimization problem. Particularly, the centre of discussion is on diagonally preconditioned limited memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) method. L-BFGS method has been widely used in large scale unconstrained optimization due to its effectiveness. However, a major drawback of the L-BFGS method is that it can be very slow on certain type of problems. Scaling and preconditioning have been used to boost the performance of the L-BFGS method. In this study, a class of diagonally preconditioned L-BFGS method will be proposed. Contrary to the standard L-BFGS method where its initial inverse Hessian approximation is the identity matrix, a class of diagonal preconditioners has been derived based upon the weak-quasi-Newton relation with an additional parameter. Choosing different parameters leads the research to some well-known diagonal updating formulae which enable the R-linear convergent for the L-BFGS method. Numerical experiments were performed on a set of large scale unconstrained minimization problem to examine the impact of each choice of parameter. The computational results suggest that the proposed diagonally preconditioned L-BFGS methods outperform the standard L-BFGS method without any preconditioning. Finally, we discuss on the impact of the diagonal preconditioners on the L-BFGS method as compared to the standard L-BFGS method in terms of the number of iterations, the number of function/gradient evaluations and the CPU time in second.
first_indexed 2025-11-15T07:30:19Z
format Thesis
id upm-7547
institution Universiti Putra Malaysia
institution_category Local University
language English
English
last_indexed 2025-11-15T07:30:19Z
publishDate 2009
recordtype eprints
repository_type Digital Repository
spelling upm-75472013-05-27T07:35:31Z http://psasir.upm.edu.my/id/eprint/7547/ A Class of Diagonally Preconditioned Limited Memory Quasi-Newton Methods for Large-Scale Unconstrained Optimization Chen, Chuei Yee The focus of this thesis is to diagonally precondition on the limited memory quasi-Newton method for large scale unconstrained optimization problem. Particularly, the centre of discussion is on diagonally preconditioned limited memory Broyden-Fletcher-Goldfarb-Shanno (L-BFGS) method. L-BFGS method has been widely used in large scale unconstrained optimization due to its effectiveness. However, a major drawback of the L-BFGS method is that it can be very slow on certain type of problems. Scaling and preconditioning have been used to boost the performance of the L-BFGS method. In this study, a class of diagonally preconditioned L-BFGS method will be proposed. Contrary to the standard L-BFGS method where its initial inverse Hessian approximation is the identity matrix, a class of diagonal preconditioners has been derived based upon the weak-quasi-Newton relation with an additional parameter. Choosing different parameters leads the research to some well-known diagonal updating formulae which enable the R-linear convergent for the L-BFGS method. Numerical experiments were performed on a set of large scale unconstrained minimization problem to examine the impact of each choice of parameter. The computational results suggest that the proposed diagonally preconditioned L-BFGS methods outperform the standard L-BFGS method without any preconditioning. Finally, we discuss on the impact of the diagonal preconditioners on the L-BFGS method as compared to the standard L-BFGS method in terms of the number of iterations, the number of function/gradient evaluations and the CPU time in second. 2009 Thesis NonPeerReviewed application/pdf en http://psasir.upm.edu.my/id/eprint/7547/1/ABS_---__FS_2009_29.pdf Chen, Chuei Yee (2009) A Class of Diagonally Preconditioned Limited Memory Quasi-Newton Methods for Large-Scale Unconstrained Optimization. Masters thesis, Universiti Putra Malaysia. Mathematical optimization - Case studies English
spellingShingle Mathematical optimization - Case studies
Chen, Chuei Yee
A Class of Diagonally Preconditioned Limited Memory Quasi-Newton Methods for Large-Scale Unconstrained Optimization
title A Class of Diagonally Preconditioned Limited Memory Quasi-Newton Methods for Large-Scale Unconstrained Optimization
title_full A Class of Diagonally Preconditioned Limited Memory Quasi-Newton Methods for Large-Scale Unconstrained Optimization
title_fullStr A Class of Diagonally Preconditioned Limited Memory Quasi-Newton Methods for Large-Scale Unconstrained Optimization
title_full_unstemmed A Class of Diagonally Preconditioned Limited Memory Quasi-Newton Methods for Large-Scale Unconstrained Optimization
title_short A Class of Diagonally Preconditioned Limited Memory Quasi-Newton Methods for Large-Scale Unconstrained Optimization
title_sort class of diagonally preconditioned limited memory quasi-newton methods for large-scale unconstrained optimization
topic Mathematical optimization - Case studies
url http://psasir.upm.edu.my/id/eprint/7547/
http://psasir.upm.edu.my/id/eprint/7547/1/ABS_---__FS_2009_29.pdf