|
Títol:
|
Web didáctica para la aproximación de funciones mediante redes neuronales
|
|
Autor/a:
|
Comabella López, Julio; Solé-Casals, Jordi; Hosseini, Sharam; Jutten, Christian
|
|
Altres autors:
|
Universitat de Vic. Escola Politècnica Superior; Universitat de Vic. Grup de Recerca en Tecnologies Digitals; Simposium de la Unión Científica Internacional de Radio (XVIIè : 2002 : Madrid ); URSI 2002 |
|
Notes:
|
In this article we presents a project [1] developed to demonstrate
the capability that Multi-Layer Perceptrons (MLP) have to
approximate non-linear functions [2]. The simulation has been
implemented in Java to be used in all the computers by Internet
[3], with a simple operation and pleasant interface. The power of
the simulations is in the possibility of the user of seeing the
evolutions of the approaches, the contribution of each neuron, the
control of the different parameters, etc. In addition, to guide the
user during the simulation, an online help has been implemented. |
|
Matèries:
|
-Web
-Xarxes neuronals (Informàtica) |
|
Drets:
|
Aquest document està subjecte a aquesta llicència Creative Commons
http://creativecommons.org/licenses/by-nc-nd/3.0/es/
|
|
Tipus de document:
|
Objecte de conferència |
|
Compartir:
|
|
