Nombre: MARCELO SOUZA FASSARELA
Tipo: MSc dissertation
Fecha de publicación: 21/12/2009
Supervisor:
Nombre |
Papel |
|---|---|
| EVANDRO OTTONI TEATINI SALLES | Advisor * |
| HANS JORG ANDREAS SCHNEEBELI | Co-advisor * |
Junta de examinadores:
| Nombre |
Papel |
|---|---|
| EVANDRO OTTONI TEATINI SALLES | Co advisor * |
| HANS JORG ANDREAS SCHNEEBELI | Advisor * |
| JOSE LEANDRO FÉLIX SALLES | Internal Examiner * |
| RENATO ANTÔNIO KROHLING | External Examiner * |
Sumario: In this work we discuss neural networks and the bias-variance dilemma.
We propose the Window method to be inserted into supervisioned neural
training with noise data. The method has an intrinsic caracteristic of regularization,
because it tries to eliminate noise while the network is beeing
trained, reducing its in uence of the adjustment of network weights. We
implement and analize the method in adaptive logic networks (ALN) and at
multilayer perceptrons (MLP). Finally, we test the network in aplications as
function aproximation, adaptive lters and time series prediction.
