Title :
The capability of feedforward neural networks with Ω-shaped activation functions
Author :
Chen, Yiwei ; Bastani, Farokh
Author_Institution :
Western Atlas Int., Houston, TX, USA
Abstract :
The capability of the multilayer feedforward layered neural network with Ω-shaped activation functions is studied. The authors prove that a three layer neural network with any continuous Ω-shaped activation function can approximate any continuous function in the multidimensional real space. Further theoretical extensions to the generalized Ω-shaped function are also explored. One example of this kind of neural network is the Hermite network
Keywords :
learning systems; neural nets; Hermite network; Omega -shaped activation functions; feedforward neural networks; multidimensional real space; Artificial neural networks; Backpropagation algorithms; Biological system modeling; Biology computing; Convergence; Feedforward neural networks; Multi-layer neural network; Neural networks; Neurons; Vectors;
Conference_Titel :
Tools for Artificial Intelligence, 1991. TAI '91., Third International Conference on
Conference_Location :
San Jose, CA
Print_ISBN :
0-8186-2300-4
DOI :
10.1109/TAI.1991.167096
