Robust prewhitening for ICA by minimizing β-divergence and its application to FastICA

Md Nurul Haque Mollah, Shinto Eguchi, Mihoko Minami

研究成果: Article

26 引用 (Scopus)

抄録

Many estimation methods for independent component analysis (ICA) requires prewhitening of observed signals. This paper proposes a new method of prewhitening named β-prewhitening by minimizing the empirical β-divergence over the space of all the Gaussian distributions. The value of the tuning parameter β plays the key role in the performance of our current proposal. An attempt is made to propose an adaptive selection procedure for the tuning parameter β for this algorithm. At last, a measure of performance index is proposed for assessing prewhitening procedures. Simulation results show that β-prewhitening efficiently improves the performance over the standard prewhitening when outliers exist; it keeps equal performance otherwise. Performance of the proposed method is compared with the standard prewhitening by both FastICA and our proposed performance index.

元の言語English
ページ(範囲)91-110
ページ数20
ジャーナルNeural Processing Letters
25
発行部数2
DOI
出版物ステータスPublished - 2007 4
外部発表Yes

Fingerprint

Independent component analysis
Tuning
Gaussian distribution
Normal Distribution

ASJC Scopus subject areas

  • Artificial Intelligence
  • Neuroscience(all)

これを引用

Robust prewhitening for ICA by minimizing β-divergence and its application to FastICA. / Mollah, Md Nurul Haque; Eguchi, Shinto; Minami, Mihoko.

:: Neural Processing Letters, 巻 25, 番号 2, 04.2007, p. 91-110.

研究成果: Article

Mollah, Md Nurul Haque ; Eguchi, Shinto ; Minami, Mihoko. / Robust prewhitening for ICA by minimizing β-divergence and its application to FastICA. :: Neural Processing Letters. 2007 ; 巻 25, 番号 2. pp. 91-110.
@article{216a93f580ca4050b2ca4d88093b5448,
title = "Robust prewhitening for ICA by minimizing β-divergence and its application to FastICA",
abstract = "Many estimation methods for independent component analysis (ICA) requires prewhitening of observed signals. This paper proposes a new method of prewhitening named β-prewhitening by minimizing the empirical β-divergence over the space of all the Gaussian distributions. The value of the tuning parameter β plays the key role in the performance of our current proposal. An attempt is made to propose an adaptive selection procedure for the tuning parameter β for this algorithm. At last, a measure of performance index is proposed for assessing prewhitening procedures. Simulation results show that β-prewhitening efficiently improves the performance over the standard prewhitening when outliers exist; it keeps equal performance otherwise. Performance of the proposed method is compared with the standard prewhitening by both FastICA and our proposed performance index.",
keywords = "β-prewhitening, Adaptive selection, Independent component analysis, One standard error, Robustness",
author = "Mollah, {Md Nurul Haque} and Shinto Eguchi and Mihoko Minami",
year = "2007",
month = "4",
doi = "10.1007/s11063-006-9023-8",
language = "English",
volume = "25",
pages = "91--110",
journal = "Neural Processing Letters",
issn = "1370-4621",
publisher = "Springer Netherlands",
number = "2",

}

TY - JOUR

T1 - Robust prewhitening for ICA by minimizing β-divergence and its application to FastICA

AU - Mollah, Md Nurul Haque

AU - Eguchi, Shinto

AU - Minami, Mihoko

PY - 2007/4

Y1 - 2007/4

N2 - Many estimation methods for independent component analysis (ICA) requires prewhitening of observed signals. This paper proposes a new method of prewhitening named β-prewhitening by minimizing the empirical β-divergence over the space of all the Gaussian distributions. The value of the tuning parameter β plays the key role in the performance of our current proposal. An attempt is made to propose an adaptive selection procedure for the tuning parameter β for this algorithm. At last, a measure of performance index is proposed for assessing prewhitening procedures. Simulation results show that β-prewhitening efficiently improves the performance over the standard prewhitening when outliers exist; it keeps equal performance otherwise. Performance of the proposed method is compared with the standard prewhitening by both FastICA and our proposed performance index.

AB - Many estimation methods for independent component analysis (ICA) requires prewhitening of observed signals. This paper proposes a new method of prewhitening named β-prewhitening by minimizing the empirical β-divergence over the space of all the Gaussian distributions. The value of the tuning parameter β plays the key role in the performance of our current proposal. An attempt is made to propose an adaptive selection procedure for the tuning parameter β for this algorithm. At last, a measure of performance index is proposed for assessing prewhitening procedures. Simulation results show that β-prewhitening efficiently improves the performance over the standard prewhitening when outliers exist; it keeps equal performance otherwise. Performance of the proposed method is compared with the standard prewhitening by both FastICA and our proposed performance index.

KW - β-prewhitening

KW - Adaptive selection

KW - Independent component analysis

KW - One standard error

KW - Robustness

UR - http://www.scopus.com/inward/record.url?scp=33847314660&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33847314660&partnerID=8YFLogxK

U2 - 10.1007/s11063-006-9023-8

DO - 10.1007/s11063-006-9023-8

M3 - Article

AN - SCOPUS:33847314660

VL - 25

SP - 91

EP - 110

JO - Neural Processing Letters

JF - Neural Processing Letters

SN - 1370-4621

IS - 2

ER -