一类新的带线搜索的自适应非单调信赖域算法

A Nonmonotone Self-adaptive Trust Region Algorithm with Line Search Based on the New Conic Model

对于无约束优化问题提出了一类基于新锥模型的带线搜索的信赖域算法。利用自适应技术,当试探步不成功时,不需重新求解子问题,而采用Armijo线搜索,以减少计算量。在适当的条件下,证明了算法的全局收敛性。

A line search trust algorithm for unconstrained optimization problems is presented. The trust radius in the new method is automatically determined with second order information based. To reduce the computational, it takes the nonmonotonic Armijo line search instead of resolving the subproblem when the trail step is not successful. Global convergence is proved under certain conditions.