A New Non-Parameter Filled Function for Global Optimization with Box Constraints and Its Application

In this paper, a new non-parameter filled function is presented to find the global minimizer of a class of nonlinear programming problems with box constraints. Based on the new filled function, the corresponding algorithm is presented. By studying the theoretical and numerical properties of the filled function and testing the algorithm on the test problems, it is proved that the method is feasible and effective, and has better global optimization ability. Furthermore, the algorithm is applied to solve the renal cell carcinoma metastasis problems, and the global optimal solution is obtained by using the filled function method to minimize the cross entropy loss function.