Title :
Compact polynomial modeling of the multi-layer perceptron
Author :
Jiang, Xianping ; Chen, Mu-Song ; Manry, Michael T.
Author_Institution :
Dept. of Electr. Eng., Texas Univ., Arlington, TX, USA
Abstract :
A technique for analyzing multilayer perceptron (MLP) neural networks is presented in which each hidden unit is modeled as a power series of the net function. This allows (1) pruning useless hidden units, (2) measuring the effect of removing a hidden layer, and (3) determining the degree of the overall polynomial discriminant (PD) which approximates the network. Classification and mapping networks are analyzed to illustrate the technique
Keywords :
neural nets; polynomials; classification networks; hidden unit; mapping networks; multi-layer perceptron; neural networks; polynomial discriminant; polynomial modeling; power series; Electric variables measurement; Equations; Measurement units; Multi-layer neural network; Multilayer perceptrons; Network topology; Neural networks; Polynomials; Power measurement; Senior members;
Conference_Titel :
Signals, Systems and Computers, 1992. 1992 Conference Record of The Twenty-Sixth Asilomar Conference on
Conference_Location :
Pacific Grove, CA
Print_ISBN :
0-8186-3160-0
DOI :
10.1109/ACSSC.1992.269164
