Name
Abstract | ||
|---|---|---|
The paper describes a methodology for constructing a possible combination of different basis functions (sigmoidal and product) for the hidden layer of a feed forward neural network, where the architecture, weights and node typology are learned based on evolutionary programming. This methodology is tested using simulated Gaussian data set classification problems with different linear correlations between input variables and different variances. It was found that combined basis functions are the more accurate for classification than pure sigmoidal or product-unit models. Combined basis functions present competitive results which are obtained using linear discriminant analysis, the best classification methodology for Gaussian data sets. |
Year | DOI | Venue |
|---|---|---|
2007 | 10.1007/978-3-540-73053-8_45 | IWINAC (1) |
Keywords | Field | DocType |
combined basis function,simulated gaussian data,different linear correlation,basis function neural networks,best classification methodology,evolutionary combining,different variance,gaussian data set,classification problem,different basis function,pure sigmoidal,linear discriminant analysis,neural network,feed forward neural network,evolutionary programming | Radial basis function network,Feedforward neural network,Pattern recognition,Computer science,Gaussian,Artificial intelligence,Basis function,Linear discriminant analysis,Artificial neural network,Evolutionary programming,Machine learning,Sigmoid function | Conference |
Volume | ISSN | Citations |
4527 | 0302-9743 | 0 |
PageRank | References | Authors |
0.34 | 5 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| César Hervás | 1 | 183 | 14.38 |
| Francisco Martínez | 2 | 0 | 0.34 |
| Mariano Carbonero-Ruz | 3 | 18 | 4.42 |
| Cristóbal Romero | 4 | 2226 | 148.97 |
| Juan Carlos Fernández | 5 | 72 | 9.77 |