In this paper, we introduce a combination of family of some conjugate gradient methods (CG) with the trust-region method. Whenever the trust-region algorithm is unsuccessful, a family of CG methods… Click to show full abstract
In this paper, we introduce a combination of family of some conjugate gradient methods (CG) with the trust-region method. Whenever the trust-region algorithm is unsuccessful, a family of CG methods is used to prevent resolving the trust-region subproblem. The computational cost for such a family is trivial. The global theory of the new approach is proved and numerical experiments are reported.
               
Click one of the above tabs to view related content.