基于新型进化规划的异构神经网络集成算法  

Algorithm of Heterogeneous Neural Network Ensemble Based on New Evolutionary Programming

作　　者：王立[1] 朱学峰[1] 

机构地区：[1]华南理工大学自动化科学与工程学院,广东广州510640

出　　处：《华南理工大学学报（自然科学版）》2009年第1期86-90,共5页Journal of South China University of Technology(Natural Science Edition)

基　　金：广东省科技攻关项目(2005B10201005)

摘　　要：为了进一步提高集成算法的泛化性能,增强个体网络生成过程的客观性,提出一种基于新型进化规划的异构神经网络集成算法.该算法首先利用改进的进化规划生成多个异构的最优网络,然后对异构网络进行组合求解.仿真实验表明,文中算法能够克服传统集成算法中成员网络结构固定、缺乏个体精度的缺点,具有比传统集成算法更好的泛化性能和更少的随机不确定因素.In order to improve the generalization ability of ensemble algorithms and the objectivity of the generating process of individual networks, an algorithm of heterogeneous neural network ensemble is proposed based on a new evolutionary programming. In this algorithm, several heterogeneous optimal neural networks are generated based on an improved evolutionary programming and are further integrated to obtain a solution. Simulated results indicate that, as compared with the traditional ensemble algorithms with fixed network structure and low individual precision, the proposed algorithm is of stronger generalization ability and fewer random elements.

关 键 词：进化规划 神经网络集成 异构神经网络 Bootstrap采样 泛化性能 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]