Article ID: | iaor2005723 |
Country: | United Kingdom |
Volume: | 31 |
Issue: | 9 |
Start Page Number: | 1411 |
End Page Number: | 1426 |
Publication Date: | Aug 2004 |
Journal: | Computers and Operations Research |
Authors: | Fang Shu-Cherng, Nuttle Henry L.W., Liao Yi |
A new neural network model is proposed based on the concepts of multi-layer perceptrons, radial basis functions, and support vector machines (SVM). This neural network model is trained using the least squared error as the optimization criterion, with the magnitudes of the weights on the links being limited to a certain range. Like the SVM model, the weight specification problem is formulated as a convex quadratic programming problem. However, unlike the SVM model, it does not require that kernel functions satisfy Mercer's condition, and it can readily extended to multi-class classification. Some experimental results are reported.