| 产生式系统与神经元网络的结合 |
| |
| 引用本文： | 王红卫,费奇.产生式系统与神经元网络的结合[J].华中科技大学学报(自然科学版),1993(6). |
| |
| 作者姓名： | 王红卫 费奇 |
| |
| 作者单位： | 华中理工大学系统工程研究所
(王红卫)，华中理工大学系统工程研究所(费奇) |
| |
| 基金项目： | 自然科学基金，“863”高技术资助项目 |
| |
| 摘 要： | 提出了产生式系统和神经元网络相结合的方法,研究了从推理网络转换到神经元网络的途径,研究了从神经元网络权矩阵转换到产生式规则的方法,从而使产生式系统能自动修正规则和获取规则.还提出了一种由二次型阈值元件组成的神经元网络,给出了它的学习算法,并严格证明了它的收敛性.
|
| 关 键 词： | 产生式系统 知识获取 推理网络 神经元网络 学习算法 |
Integration of the Production System with the Neural Network |
| |
| Wang Hongwei Fei Qi.Integration of the Production System with the Neural Network[J].JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY.NATURE SCIENCE,1993(6). |
| |
| Authors: | Wang Hongwei Fei Qi |
| |
| Institution: | Wang Hongwei Fei Qi |
| |
| Abstract: | A method of integrating the production system with the neural network is presented and the way to convert an inference network into a neural network is studied. A method for converting a neural network weight matrice into production rules is also described, so that the production system will be able to modify and acquire rules automatically. A kind of neurons composed of a quadratic form of threshold elements and its learning algorithm are proposed, and the convergence of the algorithm is strictly proved. |
| |
| Keywords: | production system knowledge acquisition inference network learning algorithm neural network |
| 本文献已被 CNKI 等数据库收录！ |
