This paper extends the use of a neural network (NN) approximating a function, to the evaluation of the gradient of the same function. This is done without any extra training of the network, The evaluation of the function's gradient is used in NN-based optimization procedures in order to speed up the convergence and to maintain the overall accuracy.
Gradient evaluation for neural networks based electromagnetic optimization procedures
ANTONINI, GIULIO;ORLANDI, Antonio
2000-01-01
Abstract
This paper extends the use of a neural network (NN) approximating a function, to the evaluation of the gradient of the same function. This is done without any extra training of the network, The evaluation of the function's gradient is used in NN-based optimization procedures in order to speed up the convergence and to maintain the overall accuracy.File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.