A modified PCA based on the minimum error entropy

Z.H. Guo, H. Yue, H. Wang

Research output: Contribution to conferencePaper

7 Citations (Scopus)

Abstract

Conventional principal component analysis (PCA) minimizes the total error variance, which may be inappropriate for the non-Gaussian distribution systems. In this paper the entropy is proposed as a more general index for PCA model, and then a modified PCA with the optimization for the minimum error entropy via a genetic algorithm (GA) is addressed.
Original languageEnglish
Pages3800-3801
Number of pages2
Publication statusPublished - Jun 2004
EventAmerican Control Conference 2004 - Boston, United States
Duration: 30 Jun 20042 Jul 2004

Conference

ConferenceAmerican Control Conference 2004
CountryUnited States
CityBoston
Period30/06/042/07/04

Fingerprint

entropy
principal component analysis
distribution system
genetic algorithm
index

Keywords

  • modified pca
  • minimum
  • error entropy
  • genetic algorithms
  • principal component analysis
  • minimum entropy methods
  • least mean squares methods

Cite this

Guo, Z. H., Yue, H., & Wang, H. (2004). A modified PCA based on the minimum error entropy. 3800-3801. Paper presented at American Control Conference 2004 , Boston, United States.
Guo, Z.H. ; Yue, H. ; Wang, H. / A modified PCA based on the minimum error entropy. Paper presented at American Control Conference 2004 , Boston, United States.2 p.
@conference{715ba441f22f40bbbc7b0262bebc6d12,
title = "A modified PCA based on the minimum error entropy",
abstract = "Conventional principal component analysis (PCA) minimizes the total error variance, which may be inappropriate for the non-Gaussian distribution systems. In this paper the entropy is proposed as a more general index for PCA model, and then a modified PCA with the optimization for the minimum error entropy via a genetic algorithm (GA) is addressed.",
keywords = "modified pca, minimum, error entropy, genetic algorithms, principal component analysis , minimum entropy methods, least mean squares methods",
author = "Z.H. Guo and H. Yue and H. Wang",
year = "2004",
month = "6",
language = "English",
pages = "3800--3801",
note = "American Control Conference 2004 ; Conference date: 30-06-2004 Through 02-07-2004",

}

Guo, ZH, Yue, H & Wang, H 2004, 'A modified PCA based on the minimum error entropy', Paper presented at American Control Conference 2004 , Boston, United States, 30/06/04 - 2/07/04 pp. 3800-3801.

A modified PCA based on the minimum error entropy. / Guo, Z.H.; Yue, H.; Wang, H.

2004. 3800-3801 Paper presented at American Control Conference 2004 , Boston, United States.

Research output: Contribution to conferencePaper

TY - CONF

T1 - A modified PCA based on the minimum error entropy

AU - Guo, Z.H.

AU - Yue, H.

AU - Wang, H.

PY - 2004/6

Y1 - 2004/6

N2 - Conventional principal component analysis (PCA) minimizes the total error variance, which may be inappropriate for the non-Gaussian distribution systems. In this paper the entropy is proposed as a more general index for PCA model, and then a modified PCA with the optimization for the minimum error entropy via a genetic algorithm (GA) is addressed.

AB - Conventional principal component analysis (PCA) minimizes the total error variance, which may be inappropriate for the non-Gaussian distribution systems. In this paper the entropy is proposed as a more general index for PCA model, and then a modified PCA with the optimization for the minimum error entropy via a genetic algorithm (GA) is addressed.

KW - modified pca

KW - minimum

KW - error entropy

KW - genetic algorithms

KW - principal component analysis

KW - minimum entropy methods

KW - least mean squares methods

M3 - Paper

SP - 3800

EP - 3801

ER -

Guo ZH, Yue H, Wang H. A modified PCA based on the minimum error entropy. 2004. Paper presented at American Control Conference 2004 , Boston, United States.