On Variable-Metric Methods for Sparse Hessians
Author:
- D. F. ShannoDSD. F. Shanno
- 1 April 1980
- journal article
- Published by JSTOR in Mathematics of Computation
- Vol. 34 (150) , 499-514
Abstract
The relationship between variable-metric methods derived by norm minimization and those derived by symmetrization of rank-one updates for sparse systems is studied, and an analogue of Dennis's nonsparse symmetrization formula derived. A new method of using norm minimization to produce a sparse analogue of any nonsparse variable-metric method is proposed. The sparse BFGS generated by this method is tested against the sparse PSB and variable-memory conjugate gradient methods, with computational experience uniformly favoring the sparse BFGS.Keywords
This publication has 17 references indexed in Scilit:
- On the use of directions of negative curvature in a modified newton methodMathematical Programming, 1979
- Least Change Secant Updates for Quasi-Newton MethodsSIAM Review, 1979
- Generating conjugate directions without line searches using factorized variable metric updating formulasMathematical Programming, 1977
- Conditioning of Quasi-Newton Methods for Function MinimizationMathematics of Computation, 1970
- A New Algorithm for Unconstrained Optimization1970
- A Family of Variable-Metric Methods Derived by Variational MeansMathematics of Computation, 1970
- Variations on variable-metric methods. (With discussion)Mathematics of Computation, 1970
- Modification of a Quasi-Newton Method for Nonlinear Equations with a Sparse JacobianMathematics of Computation, 1970
- A Class of Methods for Solving Nonlinear Simultaneous EquationsMathematics of Computation, 1965
- A Rapidly Convergent Descent Method for MinimizationThe Computer Journal, 1963