A nonmonotone trust region method based on nonincreasing technique of weighted average of the successive function values (Q2475129): Difference between revisions

The authors develop a new nonmonotone trust region method for the numerical solution of smooth unconstrained optimization problems, by combining the method proposed by \textit{H. Zhang} and \textit{W. W. Hager} [SIAM J. Optim. 14, No. 4, 1043--1056 (2004; Zbl 1073.90024)] with the trust region method. Under certain appropriate assumptions, the global convergence and also the superlinear convergence of the proposed algorithm is established. Its efficiency is then tested by numerical experiments.