Approximation of Fuzzy Neural Networks to Fuzzy-Valued Measurable Function

被引：0
|
作者：
Li, Jun [1 ]
Gao, Shuai [1 ]
Chen, Bin [1 ]
机构：
[1] Commun Univ China, Sch Sci, Beijing 100024, Peoples R China
来源：
QUANTITATIVE LOGIC AND SOFT COMPUTING 2010, VOL 2 | 2010年 / 82卷
关键词：
Fuzzy measure; Lusin's theorem; Approximation; Regular fuzzy neural network; UNIVERSAL APPROXIMATION;
D O I：
暂无
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
In this note, we shall further investigate approximation property of regular fuzzy neural network (RFNN). It is shown that any fuzzy-valued measurable function can be approximated by four-layer RFNN in the sense of Sugeno integral norm for finite weakly null-additive fuzzy measure on R. The previous result obtained by Li et al is improved.
引用
收藏
页码：351 / 358
页数：8
相关论文
共 50 条
12345