Search Results - subject:"Local error bound"

An inexact regularized proximal Newton method without line search

vom Dahl, Simeon; Kanzow, Christian - In: Computational Optimization and Applications 89 (2024) 3, pp. 585-624

In this paper, we introduce an inexact regularized proximal Newton method (IRPNM) that does not require any line search. The method is designed to minimize the sum of a twice continuously differentiable function f and a convex (possibly non-smooth and extended-valued) function φ. Instead of...

Persistent link: https://www.econbiz.de/10015361797

A new error bound result for Generalized Nash Equilibrium Problems and its algorithmic application

Dreves, Axel; Facchinei, Francisco; Fischer, Andreas; … - In: Computational Optimization and Applications 59 (2014) 1, pp. 63-84

convergence theory on a local error bound and provide a new sufficient condition for it to hold that is weaker than known ones. In …

Persistent link: https://www.econbiz.de/10010937799

A Shamanskii-like Levenberg-Marquardt method for nonlinear equations

Fan, Jinyan - In: Computational Optimization and Applications 56 (2013) 1, pp. 63-80

only a LM step but also m−1 approximate LM steps are computed, where m is a positive integer. Under the local error bound …

Persistent link: https://www.econbiz.de/10010998305

An improved trust region algorithm for nonlinear equations

Fan, Jinyan; Pan, Jianyu - In: Computational Optimization and Applications 48 (2011) 1, pp. 59-70

Persistent link: https://www.econbiz.de/10008925522

A new trust region method for nonlinear equations

Zhang, Ju-liang; Wang, Yong - In: Computational Statistics 58 (2003) 2, pp. 283-298

local error bound condition. This condition is weaker than the nondegeneracy assumption which is necessary for superlinear …

Persistent link: https://www.econbiz.de/10010847964

A new trust region method for nonlinear equations

Zhang, Ju-liang; Wang, Yong - In: Mathematical Methods of Operations Research 58 (2003) 2, pp. 283-298

local error bound condition. This condition is weaker than the nondegeneracy assumption which is necessary for superlinear …

Persistent link: https://www.econbiz.de/10010950340