C.-G. Lim, K.C. Kim, and E.-K. Kim (Korea)
adaptive wavelet network, neural network, genetic algorithm, and speech signal.
. Learning can be observed as a mapping from an input space to an output space. In this paper we present a promising approach for constructing adaptive wavelet network with single hidden layer. Usually, gradient methods such as conjugate gradient and quasi-Newton methods are used to train the network. A technique combining adaptive wavelet networks and genetic algorithm is proposed to approximate a given signal. The objective is to minimize the difference between original and approximated signals by optimizing model parameters adaptively.
Important Links:
Go Back
