학술논문

A complex valued Hebbian learning algorithm
Document Type
Conference
Source
1998 IEEE International Joint Conference on Neural Networks Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98CH36227) Neural networks Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on. 2:1235-1238 vol.2 1998
Subject
Computing and Processing
Components, Circuits, Devices and Systems
Signal Processing and Analysis
Hebbian theory
Neurons
Principal component analysis
Neural networks
Eigenvalues and eigenfunctions
Data mining
Vectors
Image coding
Radar applications
Sonar
Language
ISSN
1098-7576
Abstract
We present a training rule for a single-layered linear network with complex valued weights and activation levels. This network can be used to extract the principal components of a complex valued data set. We also introduce a new training method that reduces the training time of the complex valued as well as of the real valued network. The use of the new network and training algorithm is illustrated with a problem of compressing images represented in the spectral domain.