scientific article; zbMATH DE number 4045514
[For the entire collection see Zbl 0624.00024.] Pointwise quasi-Newton methods are iterative methods for solution of nonlinear equations. They differ from most standard quasi-Newton methods in that each is designed with a particular infinite dimensional problem in mind. In this paper we discuss the recent work on these methods and, as an example, give a new quasi-Newton method for certain systems of nonlinear integral equations. For sufficiently good initial approximations to the root and Fréchet derivative at the root the method gives superlinear convergence. For many such problems the standard approach, Broyden's method, is not guaranteed to give superlinear convergence. We close the paper with some numerical data which indicates that the convergence properties of the new method, Broyden's method, and Newton's method for the infinite dimensional problem are visible in finite dimensional approximate problems.
- Iterative solution of integral equations by a quasi-Newton method
- Mesh independence of Newton-like methods for infinite dimensional problems
- MGOPT with gradient projection method for solving bilinear elliptic optimal control problems
- A Quasi-Newton Method for Elliptic Boundary Value Problems
- scientific article; zbMATH DE number 4201491 (Why is no real title available?)
- scientific article; zbMATH DE number 4072263 (Why is no real title available?)
- A Pointwise Quasi-Newton Method for Integral Equations
- Fixed-Point Quasi-Newton Methods
- scientific article; zbMATH DE number 740991 (Why is no real title available?)
- Pointwise Broyden Methods
- Approximate quasi-Newton methods
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3783102)