Optimization of Modular Neural Network Architectures with an Improved Particle Swarm Optimization Algorithm

被引：0

作者：

Uriarte, Alfonso ^{[1
]}

Melin, Patricia ^{[1
]}

Valdez, Fevrier ^{[1
]}

机构：

[1] Tijuana Inst Technol, Div Grad Studies & Res, Tijuana, Mexico

来源：

RECENT DEVELOPMENTS AND THE NEW DIRECTION IN SOFT-COMPUTING FOUNDATIONS AND APPLICATIONS | 2018年 / 361卷

关键词：

Modular Neural Network; Particle Swarm Optimization; Pattern recognition;

D O I：

10.1007/978-3-319-75408-6_14

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

According to the literature of Particle Swarm Optimization (PSO), there are problems of getting stuck at local minima and premature convergence with this algorithm. A new algorithm is presented in this paper called the Improved Particle Swarm Optimization using the gradient descent method as an operator incorporated into the Algorithm, as a function to achieve a significant improvement. The gradient descent method (BP Algorithm) helps not only to increase the global optimization ability, but also to avoid the premature convergence problem. The Improved PSO Algorithm (IPSO) is applied to the design of Neural Networks to optimize their architecture. The results show that there is an improvement with respect to using the conventional PSO Algorithm.

引用

页码：165 / 174

页数：10