A new procedure for the sequential learning of two layer perceptrons is presented. It improves the classical method by preventing exponential growth of output weights in most cases. Single hidden units are trained by the incomplete upstart algorithm, a new promising technique which reduces the computational effort.
Sequential learning of two-layer networks with the incomplete upstart algorithm
M Muselli
1992
Abstract
A new procedure for the sequential learning of two layer perceptrons is presented. It improves the classical method by preventing exponential growth of output weights in most cases. Single hidden units are trained by the incomplete upstart algorithm, a new promising technique which reduces the computational effort.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
