折线模糊神经网络对Chonquet可积函数的泛逼近性
Approximation of Polygonal Fuzzy Neural Networks for Chonquet Integrable Fuzzy Functions
-
摘要: 首先定义了Choquet积分模的概念,然后分析了折线模糊神经网络在Choquet积分模意义下对模糊值函数的泛逼近性,证明了当模糊值函数满足相容性时,折线模糊神经网络能够以任意精度逼近该Choquet可积模糊值函数.Abstract: First the Choquet integral norms in sub-additive fuzzy measure are introduced in this paper.Then the universal approximation of polygonal fuzzy neural networks for fuzzy valued functions in sense of Choquet integral norm is analyzed.We show that the polygonal fuzzy neural networks are pan-approximators for fuzzy-valued functions in sense of Choquet integral norms with respect to fuzzy measure when the fuzzy-valued functions are compatible.
下载:
