Title :
Sizing of the multilayer perceptron via modular networks
Author :
Chandrasekaran, Hema ; Kim, Kyung K. ; Manry, Michael T.
Author_Institution :
Dept. of Electr. Eng., Texas Univ., Arlington, TX, USA
Abstract :
A fast method for sizing the multilayer perceptron is proposed. The principal assumption is that a modular network with the same theoretical pattern storage as the multilayer perceptron has the same training error. This assumption is analyzed for the case of random patterns. Using several benchmark datasets, the validity of the approach is demonstrated
Keywords :
convergence; learning (artificial intelligence); multilayer perceptrons; probability; modular networks; random patterns; sizing method; theoretical pattern storage; training error; Aircraft propulsion; Clustering algorithms; Convergence; Electronic mail; Linear approximation; Multilayer perceptrons; Pattern analysis; Piecewise linear approximation; Piecewise linear techniques; Vectors;
Conference_Titel :
Neural Networks for Signal Processing IX, 1999. Proceedings of the 1999 IEEE Signal Processing Society Workshop.
Conference_Location :
Madison, WI
Print_ISBN :
0-7803-5673-X
DOI :
10.1109/NNSP.1999.788140
