Convergence of inexact Newton methods for generalized equations

Variational inequalities (49J40) Set-valued and variational analysis (49J53) Sensitivity, stability, well-posedness (49K40) Newton-type methods (49M15) Numerical methods for variational inequalities and related problems (65K15) Nonlinear programming (90C30)

Review text

For general inclusions of the form ``zero is contained in \(f(x) + F(x)\), where \(f\) is a smooth function and \(F\) a set-valued mapping both acting between Banach spaces, the authors study local properties of inexact Newton methods. As main results they get conditions such that the considered iteration sequences have ``no halt, that means they are surely executable, and about the quality of their convergence (e.g., q-linear, q-quadratic, q-superlinear). After carefully explaining kinds of metric regularity and relations to connected topics as Aubin continuity, calmness, inverse function theorem or Ljusternik-Graves theorem and after proving some interesting theoretical results (e.g., about perturbed metric regularity), several convergence theorems are proved. Miscellaneous examples and counterexamples follow in a concluding section together with various hints to former results in the literature.

Recommendations

Cites work

Cited in

(45)

This page was built for publication: Convergence of inexact Newton methods for generalized equations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q353160)