I. Magdisiuk et A. Borisov, Cascade-correlated architecture of neural network in problems of classification and prediction, AUTOM C C S, 34(4), 2000, pp. 27-33
Features of the application of the cascade-correlated architecture of a neu
ral network as well as the potential for adaptation of this type of archite
cture to the solution of problems of classification and prediction are stud
ied. The effect of the method of interpretation of the output signal of the
network and the influence of different combinations of methods of correcti
on of the error on the convergence of the network is studied.