A dense initialization for limited-memory quasi-Newton methods

DOI10.1007/s10589-019-00112-xzbMath1427.90292arXiv1710.02396MaRDI QIDQ2322555

Oleg P. Burdakov, Jennifer B. Erway, Roummel F. Marcia, Johannes J. Brust

Publication date: 4 September 2019

Published in: Computational Optimization and Applications (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/1710.02396

zbMATH Keywords

trust-region methods; limited-memory quasi-Newton methods; large-scale nonlinear optimization; quasi-Newton matrices; shape-changing norm

Mathematics Subject Classification ID

65F15: Numerical computation of eigenvalues and eigenvectors of matrices

65K05: Numerical mathematical programming methods

90C06: Large-scale problems in mathematical programming

90C26: Nonconvex programming, global optimization

65K10: Numerical optimization and variational techniques

90C53: Methods of quasi-Newton type

65F10: Iterative numerical methods for linear systems

Uses Software

• L-BFGS-B
• L-BFGS
• CUTEr
• SifDec
• LBFGS-B