一类新的信赖域算法的全局收敛性

本文对于无约束最优化问题提出了一类非单调的信赖域算法，它是通常的单调信赖域算法的推广。当目标函数是有下界的连续可微函数，而且它的二阶导数的近似的模是线性地依赖于迭代次数时，我们证明了新算法的整体收敛性。

GLOBAL CONVERGENCE OF A CLASS OF NEW TRUST REGION ALGORITHMS

In this paper we propose a class of nonmonotone trust region algorithms for unconstrained optimization. When the objective function is bounded below and continuously differentiable, and there is a bound on the second derivative approximations that depends linearly on the iteration number, we prove the global convergence of the algorithms.