Download:
|
by Michael Drexler, Gene H. Golub
SIAM Journal on Numerical Analysis
http://www-sccm.stanford.edu/~drexler/docs/Prox_Theo.ps.Z
Add To MetaCart
Abstract:
We analyse the elementwise convergence in the Jacobian of the classical Newton method as the solution of a non-linear system f (x) = 0 is approached. It is shown that each Jacobian element approaches a constant value at a rate determined by the nonlinearity of the relevant function with respect to the appropriate scalar variable. For many practical problems, this means that in the last steps of the Newton iteration, only a low-rank update has to be applied to the Jacobian. The consequences of these results are examined for both direct and iterative solution methods, and appropriate algorithmic modifications stated. Finally, two numerical examples are presented to illustrate the theory.
Citations
|
599
|
Numerical Methods for Unconstrained Optimization and Nonlinear Equations, Prentice-Hall
– DENNIS, SCHNABEL
- 1983
|
|
352
|
Rheinboldt, Iterative Solution of Nonlinear Equations
– Ortega, W
- 2000
|
|
175
|
Inexact Newton methods
– DEMBO, EISENSTAT, et al.
- 1982
|
|
83
|
The differentiation of pseudoinverses and nonlinear least squares problems whose variables separate
– Golub, Pereyra
- 1973
|
|
76
|
Methods for Modifying Matrix Factorizations
– Golub, Murray, et al.
- 1974
|
|
60
|
Choosing the forcing terms in an inexact Newton method
– EISENSTAT, WALKER
- 1996
|
|
45
|
Numerical Solution of Partial Differential Equations
– Morton, Mayers
- 1994
|
|
42
|
Quasi-Newton methods, motivation and theory
– Dennis, More
- 1977
|
|
40
|
Newton’s Method Estimates from Data at One Point in: The Merging of Disciplines
– Smale
- 1986
|
|
28
|
Global approximate Newton methods
– Bank, Rose
- 1981
|
|
8
|
On algorithms for solving f(x
– Hirsch, Smale
- 1979
|
|
5
|
Templates for the Solution of Linear Systems: Building Blocks for Iterative Methods
– al
- 1994
|