A Quasi-Newton Quadratic Penalty Method for Minimization Subject toNonlinear Equality Constraints

Authors:
Thomas F. Coleman;Jianguo Liu;Wei Yuan
Affiliations:
Computer Science Department and Cornell Theory Center, Cornell University, Ithaca, NY 14850, USA;Department of Mathematics, University of North Texas, Denton, TX 76203, USA;Center for Applied Mathematics, Cornell University, Ithaca, NY 14853, USA
Venue:
Computational Optimization and Applications
Year:
2000

Citing 7
Cited 1

Continuity of the null space basis and constrained optimization

Mathematical Programming: Series A and B
Local convergence of secant methods for nonlinear constrained optimization

SIAM Journal on Numerical Analysis
On the convegence of a sequential penalty function method for constrained minimization

SIAM Journal on Numerical Analysis
Computing a trust region step for a penalty function

SIAM Journal on Scientific and Statistical Computing
An analysis of reduced Hessian methods for constrained optimization

Mathematical Programming: Series A and B
Analysis and implementation of a dual algorithm for constrained optimization

Journal of Optimization Theory and Applications
Numerical Methods for Unconstrained Optimization and Nonlinear Equations (Classics in Applied Mathematics, 16)

Numerical Methods for Unconstrained Optimization and Nonlinear Equations (Classics in Applied Mathematics, 16)

A New Trust-Region Algorithm for Equality Constrained Optimization

Computational Optimization and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present a modified quadratic penalty function method forequality constrained optimization problems. The pivotal feature ofour algorithm is that at every iterate we invoke a special change ofvariables to improve the ability of the algorithm to follow theconstraint level sets. This change of variables gives rise to asuitable block diagonal approximation to the Hessian which is thenused to construct a quasi-Newton method. We show that the completealgorithm is globally convergent. Preliminary computational resultsare reported.