文献类型：期刊文献

中文题名：基于多输入泛函网络的构造和学习策略

英文题名：Strategy of Structuring and Learning Based on Multi-Input and Single Output Functional Network

作者：崔明义[1];张新祥[1];苏白云[1];张瑞[1]

第一作者：崔明义

机构：[1]河南财经学院计算机科学系

第一机构：河南财经政法大学计算机与信息工程学院

年份：2006

卷号：33

期号：10

起止页码：169-171

中文期刊名：计算机科学

外文期刊名：Computer Science

收录：CSTPCD;;北大核心:【北大核心2004】；CSCD:【CSCD2011_2012】；

基金：河南省自然科学基金(0411013800;0411014500);河南省高校杰出科研人才创新工程项目(2004KYCX014)资助

语种：中文

中文关键词：泛函网络;MIOFN;泛函方程;拓扑结构;学习

外文关键词：Functional network, MIOFN,Functional equation, Topology structure, Learning

摘要：泛函网络是类似于人工神经网络的新型网络模型,是泛函方程的网络表达形式。本文针对复杂泛函网络构造和学习中存在的问题,提出了多输入泛函网络模型MIOFN。通过对该模型的分析,提出了简化和学习的方法,并进行了仿真实验。结果表明,本文提出的MIOFN运行是可靠的,在工程应用中是有效可行的。
Functional network is new network model. It is similar to artificial neural network and is network expression of functional equation. In this paper, a multi-input and single output FN were presented for solving problems in structuring and learning of complex FN. Through the model was analyzed, a new simplifying and learning method was put forward by it. The Simulating experiment was done by it. The result indicates that running of the model is reliable. It is applied to engineering availably and feasibly.