一种基于协同进化的FNN结构优化和集成

Structure Optimization and Ensemble of FNN Based on Cooperative Coevolution

FNN的结构优化和集成是ANN研究的重要课题,用遗传进化的方法是解决该问题的重要途径,论文基于协同进化的理论,用多目标优化的思想,对该问题进行了研究,提出了解决该问题的方法,在理论分析的基础上进行了仿真实验,结果证明,该方法理论上是可靠的,实践上是可行的。

The structure optimization and ensemble is an important issue of ANN research. Genetic evolution is an important approach of solving the problem.This paper discueses the issue,with.algorithms of multi-objective optimization and theory of cooperative coevolution.The method is proposed to solve the problem.Simulation experiment is carried out on the theory analysis.The reliability of theory and feasibility of practice are proved by the result in it.