Title :
Principal components via cascades of block-layers
Author :
Palmieri, Francesco ; Corvino, Michele
Author_Institution :
Dipartimento di Ingegneria Elettronica, Naples Univ., Italy
Abstract :
We study the behaviour of the cascade of linear neural networks having constrained connectivity of Hebbian and anti-Hebbian synapses. We derive a formula for the number of layers necessary to obtain a sufficiently close approximation to the principal components at the final output. Results of simulations confirm the analyses
Keywords :
Hebbian learning; multilayer perceptrons; Hebbian synapses; anti-Hebbian synapses; block-layer cascades; constrained connectivity; linear neural network cascade; principal components; Adaptive signal processing; Analytical models; Computational modeling; Computer architecture; Computer networks; Decorrelation; Hardware; Image analysis; Jacobian matrices; Neural networks;
Conference_Titel :
Neural Networks,1997., International Conference on
Conference_Location :
Houston, TX
Print_ISBN :
0-7803-4122-8
DOI :
10.1109/ICNN.1997.616170
