Generalizing Geometric ICA to Nonlinear Settings

Authors:
Fabian J. Theis;Carlos G. Puntonet;Elmar W. Lang
Affiliations:
Institute of Biophysics, AG Neuro- and Bioinformatics, University of Regensburg, Regensburg, Germany D-93040;Dept. Arquitecura y Tecnologia de Computadores, Escuela Tcnica Superior de Ingenieria Informatica, Universidad de Granada, Granada E-18071;Institute of Biophysics, AG Neuro- and Bioinformatics, University of Regensburg, Regensburg, Germany D-93040
Venue:
IWANN '03 Proceedings of the 7th International Work-Conference on Artificial and Natural Neural Networks: Part II: Artificial Neural Nets Problem Solving Methods
Year:
2009

Citing 4
Cited 0

Space or time adaptive signal processing by neural network models

AIP Conference Proceedings 151 on Neural Networks for Computing
An application of the principle of maximum information preservation to linear systems

Advances in neural information processing systems 1
Linear geometric ICA: fundamentals and algorithms

Neural Computation
Source separation in post-nonlinear mixtures

IEEE Transactions on Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, a geometry-based algorithm for nonlinear blind source separation is presented. The mixture space is decomposed in a set of concentric rings, in which ordinary linear ICA is performed in order to get a set of images of ring points under the original mixing mapping. Putting those together the mixing mapping can be reconstructed. Various applications to two- and three-dimensional artificial and natural data sets are presented.