[1]薄迎春,乔俊飞,杨刚.一种多模块协同参与的神经网络[J].智能系统学报,2011,6(3):225-230.
BO Yingchun,QIAO Junfei,YANG Gang.A multimodule cooperative neural network[J].CAAI Transactions on Intelligent Systems,2011,6(3):225-230.
点击复制
《智能系统学报》[ISSN 1673-4785/CN 23-1538/TP] 卷:
6
期数:
2011年第3期
页码:
225-230
栏目:
学术论文—机器学习
出版日期:
2011-06-25
- Title:
-
A multimodule cooperative neural network
- 文章编号:
-
1673-4785(2011)03-0225-06
- 作者:
-
薄迎春,乔俊飞,杨刚
-
北京工业大学 电子信息与控制工程学院,北京 100124
- Author(s):
-
BO Yingchun, QIAO Junfei, YANG Gang
-
College of Electronic Information and Control Engineering, Beijing University of Technology, Beijing 100124, China
-
- 关键词:
-
神经网络; 条件模糊聚类; 多模块; 子网络选择
- Keywords:
-
neural networks; conditional fuzzy clustering method; multimodules; subnets selection
- 分类号:
-
TP183
- 文献标志码:
-
A
- 摘要:
-
针对单一神经网络训练时间长、对复杂问题处理精度较低、易陷入局部极小等问题,提出了一种多模块协同参与信息处理的神经网络.该神经网络具有层级结构,基于条件模糊聚类技术对样本进行分类,根据分类结果实现对神经网络的模块划分,采用代数算法对网络权值进行求解,基于距离测度设计了处理输入信息的子网络选择方法.为提高神经网络对复杂问题的逼近能力,选择数目不等的多个子网络参与给定输入的协同处理,采取“分而治之”与“集成学习”相结合方法以提高网络的性能.实验表明,对于复杂问题,这种多模块协同参与的神经网络可以有效地提高网络的逼近精度,训练时间也优于单一网络.
- Abstract:
-
Aiming to solve the problems of long training time, low precision in processing complex problem, and a local minimum in single neural networks, a multimodule cooperative neural network (MMCNN) was proposed. Its structure has hierarchical character. Sample data was first detached by the fuzzy clustering method, and then the neural network was partitioned into several subnets based on the clustering results. The linking weights were elicited by solving equations. For a given input data, some multimodules were selected to deal with it. The approximating performance was improved by combining divideandconquer and learning ensemble strategies. A subnet selection method was designed based on distance measurements. Simulation results demonstrate that a multimodule cooperative neural network can heighten approximating ability effectively for complicated problems, and the training time is faster than in a single backpropagation neural network.
备注/Memo
收稿日期:2010-04-07.
基金项目:国家自然科学基金资助项目(60873043);北京市自然科学基金资助项目(4092010).
通信作者:薄迎春.E-mail:boyingchun@sina.com.
作者简介:
薄迎春,男,1977年生,博士研究生,讲师,主要研究方向为神经计算与智能优化控制.
乔俊飞,男,1968年生,教授,博士生导师,博士,中国人工智能学会理事、科普工作委员会秘书长,中国自动化学会智能专业委员会委员,北京自动化学会常务理事,北京人工智能学会理事、秘书长.主要研究方向为复杂过程建模与控制、神经计算与智能优化控制.主持多项国家“863”计划项目,发表学术论文多篇,被SCI、EI检索50余篇.
杨刚,男,1983年生,博士研究生,主要研究方向为神经计算与智能优化控制.
更新日期/Last Update:
2011-07-23