Neural network implementations of independent component analysis

Author

Mutihac, Radu ; Van Hulle, Marc M.

Author_Institution

Lab. voor Neuro- en Psychofysiologie, Katholieke Univ., Leuven, Belgium

fYear

2002

fDate

2002

Firstpage

505

Lastpage

514

Abstract

The performance of six neuromorphic adaptive structurally different algorithms was analyzed in blind separation of independent artificially generated signals using the stationary linear independent component analysis (ICA) model. The estimated independent components were assessed and compared aiming to rank the neural ICA implementations. All algorithms were run with different contrast functions, which were optimally selected on the basis of maximizing the sum of individual negentropies of the network outputs. Both subGaussian and superGaussian one-dimensional time series were employed throughout the numerical simulations.

Keywords

Gaussian processes; adaptive signal processing; blind source separation; feedforward neural nets; independent component analysis; time series; 1D time series; blind separation; contrast functions; feedforward neural network architecture; independent artificially generated signals; independent component analysis; network output; neural network implementations; neuromorphic adaptive algorithms; numerical simulations; stationary linear ICA model; subGaussian one-dimensional time series; superGaussian one-dimensional time series; Array signal processing; Artificial neural networks; Independent component analysis; Neural networks; Neuromorphics; Principal component analysis; Psychology; Signal processing algorithms; Source separation; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks for Signal Processing, 2002. Proceedings of the 2002 12th IEEE Workshop on

Print_ISBN

0-7803-7616-1

Type

conf

DOI

10.1109/NNSP.2002.1030062

Filename

1030062