The convergence of variable metric matrices in unconstrained optimization

From MaRDI portal
Publication:3314451





The authors address the question of convergence of the sequence of variable metric matrices for certain quasi-Newton algoritms in unconstrained optimization. It is assumed that the objective function has a Lipschitz continuous derivative and a positive definite Hessian at the limit point of the iterates. Then, for the DFP- and BFGS-method with unit step length, the sequence of update matrices converges to a limit.




Cited in
(23)








This page was built for publication: The convergence of variable metric matrices in unconstrained optimization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3314451)