Linear and nonlinear ICA based on mutual information

Author

Almeida, Luis B.

Author_Institution

INESC, Lisbon, Portugal

fYear

2000

fDate

2000

Firstpage

117

Lastpage

122

Abstract

In the context of independent components analysis (ICA), the mutual information (MI) of the extracted components is one of the most desirable measures of independence, due to its special properties. This paper presents a method for performing linear and nonlinear ICA based on MI, with few approximations. The use of MI as an objective function for ICA requires the estimation of the statistical distributions of the separated components. In this work, both the extraction of independent components and the estimation of their distributions are performed simultaneously, by a single network with a specialized structure, trained with a single objective function

Keywords

information theory; learning (artificial intelligence); multilayer perceptrons; statistical analysis; approximations; extracted components; extraction components; independence measure; independent components analysis; linear ICA; multilayer perceptrons; mutual information; nonlinear ICA; objective function; statistical distribution estimation; trained network; Blind source separation; Data analysis; Data mining; Ear; Independent component analysis; Mutual information; Performance evaluation; Source separation; Statistical distributions; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Adaptive Systems for Signal Processing, Communications, and Control Symposium 2000. AS-SPCC. The IEEE 2000

Conference_Location

Lake Louise, Alta.

Print_ISBN

0-7803-5800-7

Type

conf

DOI

10.1109/ASSPCC.2000.882457

Filename

882457