Title | ||
---|---|---|
Accelerating the convergence speed of neural networks learning methods using least squares |
Abstract | ||
---|---|---|
In this work a hybrid training scheme for the supervised learning of feedforward neural networks is presented. In the proposed method, the weights of the last layer are obtained employing linear least squares while the weights of the previous layers are updated using a stan- dard learning method. The goal of this hybrid method is to assist the ex- isting learning algorithms in accelerating their convergence. Simulations performed on two data sets show that the proposed method outperforms, in terms of convergence speed, the Levenberg-Marquardt algorithm. |
Year | Venue | Keywords |
---|---|---|
2003 | ESANN | neural network,feedforward neural network,supervised learning,levenberg marquardt,least square |
Field | DocType | Citations |
Online machine learning,Least squares support vector machine,Pattern recognition,Computer science,Wake-sleep algorithm,Supervised learning,Multilayer perceptron,Types of artificial neural networks,Artificial intelligence,Deep learning,Artificial neural network,Machine learning | Conference | 9 |
PageRank | References | Authors |
0.71 | 8 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Oscar Fontenla-Romero | 1 | 337 | 39.49 |
Deniz Erdogmus | 2 | 1299 | 169.92 |
José Carlos Príncipe | 3 | 841 | 102.43 |
Amparo Alonso-Betanzos | 4 | 885 | 76.98 |
Enrique Castillo | 5 | 555 | 59.86 |