Limited-memory variable metric methods that use quantities from the preceding iteration.

MaRDI QIDQ3456431zbMATH OpenFDO

Authors Jan Vlček, Ladislav Lukšan

Publication date 14 December 2015

Full work available at URL http://hdl.handle.net/10338.dmlcz/702762

zbMATH Keywords

algorithmsglobal convergenceL-BFGS methodlimited-memory methodsnumerical resultsquasi-Newton methodsStrang recurrencesunconstrained optimizationvariable metric methods

Mathematics Subject Classification ID

Numerical mathematical programming methods (65K05) Nonlinear programming (90C30) Methods of quasi-Newton type (90C53)

Recommendations

• Generalizations of the limited-memory BFGS method based on the quasi-product form of update
• Modifications of the limited-memory BFGS method based on the idea of conjugate directions.
• Shifted limited-memory variable metric methods for large-scale unconstrained optimization
• scientific article; zbMATH DE number 1292995
• A limited-memory optimization method using the infinitely many times repeated BNS update and conjugate directions

Cited in

(3)

• Adding variables to quasi-newton Hessian approximations
• Generalizations of the limited-memory BFGS method based on the quasi-product form of update
• scientific article; zbMATH DE number 6176035 (Why is no real title available?)

This page was built for publication: Limited-memory variable metric methods that use quantities from the preceding iteration.

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3456431)