Abstract | ||
---|---|---|
Neural networks are widely used to approximate nonlinear functions. In order to study its approximation capability, a theorem of integral representation of functions is developed by using integral transform. Using the developed representation, an approximation order estimation for the bell-shaped neural networks is obtained. The obtained result reveals that the approximation accurately of the bell-shaped neural networks depends not only on the number of hidden neurons, but also on the smoothness of target functions. |
Year | DOI | Venue |
---|---|---|
2006 | 10.1007/11759966_10 | ISNN (1) |
Keywords | Field | DocType |
approximation capability,network approximation,neural network,target function,hidden neuron,developed representation,approximate nonlinear function,bell-shaped neural network,integral representation,approximation order estimation,integral transforms | Applied mathematics,Nonlinear system,Pattern recognition,Computer science,Mathematical analysis,Integral representation,Artificial intelligence,Smoothness,Artificial neural network,Integral transform,Function representation,Sigmoid function | Conference |
Volume | ISSN | ISBN |
3971 | 0302-9743 | 3-540-34439-X |
Citations | PageRank | References |
0 | 0.34 | 7 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Fengjun Li | 1 | 233 | 23.55 |
Zongben Xu | 2 | 3203 | 198.88 |