MISEP - An ICA method for linear and nonlinear mixtures, based on mutual information

Proceedings of the International Joint Conference on Neural Networks

Volumn 1, Issue , 2002, Pages 442-447

  Almeida, Luís B ^a  

^a INESC ID   (Portugal)

Author keywords

[No Author keywords available]

Indexed keywords

COMPUTER SIMULATION; FUNCTIONS; ITERATIVE METHODS; MULTILAYER NEURAL NETWORKS; OPTIMIZATION; PROBABILITY DISTRIBUTIONS; SIGNAL TO NOISE RATIO;

CUMULATIVE PROBABILITY FUNCTION; GRADIENT-BASED OPTIMIZATION; MUTUAL INFORMATION; OBJECTIVE FUNCTION; SOFTWARE PACKAGE MATLAB;

INDEPENDENT COMPONENT ANALYSIS;

EID: 0036079634     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (18)

1. An unifying information-theoretic framework for independent component analysis
2. Blind separation of sources: A nonlinear neural algorithm
3. Nonlinear higher-order statistical decorrelation by volume-conserving neural architectures
4. An objective function for independence
5. Separation of nonlinear mixtures using pattern repulsion
6. Nonlinear independent component analysis using ensemble learning: Theory
7. Linear and nonlinear ICA based on mutual information
8. Independent component analysis - A new concept?
9. Jacobi angles for simultaneous diagonalization
10. A fast fixed-point algorithm for independent component analysis
11. Simultaneous MI-based estimation of independent components and of their distributions
12. A new learning algorithm for blind signal separation
13. A new activation function for blind signal separation
14. An information-maximization approach to blind separation and blind deconvolution
15. Entropy optimization - Application to blind separation of sources
16. Analyse générale des liaisons stochastiques
17. Nonlinear independent component analysis: Existence and uniqueness results
18. Multilayer perceptrons