大连工业大学学报2012，Vol.31Issue(5)：376-378,3.
带递归单元的模糊感知器的δ-规则的有限收敛性
Finite convergence of δ-rule for a recurrent fuzzy perceptron
刘燕 1阎慧臻 1刘超 1姜行健 2杨开兵1
作者信息
- 1. 大连工业大学信息科学与工程学院,辽宁大连 116034
- 2. 大连理工大学机械工程与材料能源学部,辽宁大连 16024
- 折叠
摘要
Abstract
A fuzzy δ-rule training algorithm was proposed for a recurrent fuzzy perceptron, in which the training patterns were supplied in completely stochastic order. It was proved that fuzzy δ-rule training algorithm was finitely convergent in the case that the training patterns were fuzzily separable. The training patterns could be correctly classified by the net after finite steps of iterating so that the perceptron could effectively solve classification problems.关键词
递归/模糊感知器/模糊可分Key words
recurrent/ fuzzy perceptron/ fuzzily separable分类
信息技术与安全科学引用本文复制引用
刘燕,阎慧臻,刘超,姜行健,杨开兵..带递归单元的模糊感知器的δ-规则的有限收敛性[J].大连工业大学学报,2012,31(5):376-378,3.
