A globally convergent trust-region algorithm for unconstrained derivative-free optimization
نویسندگان
چکیده
In this work we explicit a derivative-free trust-region algorithm for unconstrained optimization based on the paper (Computational Optimization and Applications 53: 527–555, 2012) proposed by Powell. The objective function is approximated by quadratic models obtained by polynomial interpolation. The number of points of the interpolation set is fixed. In each iteration only one interpolation point can be replaced and consequently the objective function is evaluated only once per iteration. The method involves two types of steps: trustregion and alternative steps. In a trust-region step the model is minimized in the hope that the reduction achieved by the model is inherited by the objective function. An alternative step aims to keep the good position of the interpolation points. We present the method in an algorithmic scheme and we discuss in details the results related to the global convergence of the algorithm, proving that, under reasonable hypotheses, any accumulation point of the sequence generated by the algorithm is stationary.
منابع مشابه
A Trust-region Method using Extended Nonmonotone Technique for Unconstrained Optimization
In this paper, we present a nonmonotone trust-region algorithm for unconstrained optimization. We first introduce a variant of the nonmonotone strategy proposed by Ahookhosh and Amini cite{AhA 01} and incorporate it into the trust-region framework to construct a more efficient approach. Our new nonmonotone strategy combines the current function value with the maximum function values in some pri...
متن کاملA new trust-region algorithm based on radial basis function interpolation
Optimization using radial basis functions as an interpolation tool in trust-region (ORBIT), is a derivative-free framework based on fully linear models to solve unconstrained local optimization, especially when the function evaluations are computationally expensive. This algorithm stores the interpolation points and function values to using at subsequent iterations. Despite the comparatively ad...
متن کاملA Globally Convergent Primal-dual Interior-point Filter Method for Nonlinear Programming: New Filter Optimality Measures and Computational Results
In this paper we prove global convergence for first and second-order stationarity points of a class of derivative-free trust-region methods for unconstrained optimization. These methods are based on the sequential minimization of linear or quadratic models built from evaluating the objective function at sample sets. The derivative-free models are required to satisfy Taylor-type bounds but, apar...
متن کاملA Filter-Trust-Region Method for Unconstrained Optimization
A new filter-trust-region algorithm for solving unconstrained nonlinear optimization problems is introduced. Based on the filter technique introduced by Fletcher and Leyffer, it extends an existing technique of Gould, Leyffer, and Toint [SIAM J. Optim., 15 (2004), pp. 17–38] for nonlinear equations and nonlinear least-squares to the fully general unconstrained optimization problem. The new algo...
متن کاملSolving the Unconstrained Optimization Problems Using the Combination of Nonmonotone Trust Region Algorithm and Filter Technique
In this paper, we propose a new nonmonotone adaptive trust region method for solving unconstrained optimization problems that is equipped with the filter technique. In the proposed method, the various nonmonotone technique is used. Using this technique, the algorithm can advantage from nonmonotone properties and it can increase the rate of solving the problems. Also, the filter that is used in...
متن کامل