A nonmonotone trust region method with new inexact line search for unconstrained optimization (Q369443)

The authors introduce a nonmonotone rule and incorporate the nonmonotone strategy into the trust region method for solving unconstrained optimization problems. The proposed algorithm does not restrict the objective function values to be monotonically decreasing. The algorithm combines line search to find an iterative point instead of resolving the trust region subproblems. The global convergence of the algorithm is established under certain conditions. Some numerical results are presented to show the efficiency of the algorithm.