Catalogo Articoli (Spogli Riviste)

OPAC HELP

Titolo:
Multimodal independent component analysis - A method of feature extractionfrom multiple information sources
Autore:
Akaho, S; Umeyama, S;
Indirizzi:
Electrotech Lab, Tsukuba, Ibaraki 3058568, Japan Electrotech Lab Tsukuba Ibaraki Japan 3058568 uba, Ibaraki 3058568, Japan
Titolo Testata:
ELECTRONICS AND COMMUNICATIONS IN JAPAN PART III-FUNDAMENTAL ELECTRONIC SCIENCE
fascicolo: 11, volume: 84, anno: 2001,
pagine: 21 - 28
SICI:
1042-0967(2001)84:11<21:MICA-A>2.0.ZU;2-C
Fonte:
ISI
Lingua:
ENG
Soggetto:
MUTUAL INFORMATION;
Keywords:
multimodal data; mutual information; canonical correlation analysis; independent component analysis; Gram-Charlier expansion; natural gradient;
Tipo documento:
Article
Natura:
Periodico
Settore Disciplinare:
Engineering, Computing & Technology
--discip_EC--
Citazioni:
15
Recensione:
Indirizzi per estratti:
Indirizzo: Akaho, S Electrotech Lab, 1-1-4 Umezono, Tsukuba, Ibaraki 3058568, Japan Electrotech Lab 1-1-4 Umezono Tsukuba Ibaraki Japan 3058568 Japan
Citazione:
S. Akaho e S. Umeyama, "Multimodal independent component analysis - A method of feature extractionfrom multiple information sources", ELEC C JP 3, 84(11), 2001, pp. 21-28

Abstract

We propose a method to extract features from a pair of multivariate information sources. CCA (canonical correlation analysis) is a traditional methodfor this purpose, but it does not always succeed in producing features that are nonlinearly related, because of a Gaussian assumption. We extend the framework of CCA by introducing a criterion used in ICA (independent component analysis). The proposed framework MICA (multimodal ICA) maximizes mutual information between a pair of extracted features from the two modalities,with the features extracted from each modality being statistically independent. The cost function is given by a weighted sum of the two criteria, andit is approximated by Gram-Charlier expansion. The gradient descent learning algorithm is derived to optimize this approximated function by taking a natural gradient. (C) 2001 Scripta Technica.

ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 31/03/20 alle ore 19:31:51