A fast fixed-point algorithm for complexity pursuit

Zhenwei Shi; Huanwen Tang; Yiyuan Tang

doi:10.1016/j.neucom.2004.11.028

A fast fixed-point algorithm for complexity pursuit

Zhenwei Shi, Huanwen Tang, Yiyuan Tang

Psychological Sciences

Research output: Contribution to journal › Article › peer-review

30 Scopus citations

Abstract

Complexity pursuit is a recently developed algorithm using the gradient descent for separating interesting components from time series. It is an extension of projection pursuit to time series data and the method is closely related to blind separation of time-dependent source signals and independent component analysis (ICA). In this paper, a fixed-point algorithm for complexity pursuit is introduced. The fixed-point algorithm inherits the advantages of the well-known FastICA algorithm in ICA, which is very simple, converges fast, and does not need choose any learning step sizes.

Original language	English
Pages (from-to)	529-536
Number of pages	8
Journal	Neurocomputing
Volume	64
Issue number	1-4 SPEC. ISS.
DOIs	https://doi.org/10.1016/j.neucom.2004.11.028
State	Published - Mar 2005

Keywords

Blind source separation
Complexity pursuit
Independent component analysis
Projection pursuit
Time series

Access to Document

10.1016/j.neucom.2004.11.028

Cite this

@article{bf6e7cf4ba5546c492f998f419c3f55c,

title = "A fast fixed-point algorithm for complexity pursuit",

abstract = "Complexity pursuit is a recently developed algorithm using the gradient descent for separating interesting components from time series. It is an extension of projection pursuit to time series data and the method is closely related to blind separation of time-dependent source signals and independent component analysis (ICA). In this paper, a fixed-point algorithm for complexity pursuit is introduced. The fixed-point algorithm inherits the advantages of the well-known FastICA algorithm in ICA, which is very simple, converges fast, and does not need choose any learning step sizes.",

keywords = "Blind source separation, Complexity pursuit, Independent component analysis, Projection pursuit, Time series",

author = "Zhenwei Shi and Huanwen Tang and Yiyuan Tang",

note = "Funding Information: We are grateful to all the anonymous reviewers who provided insightful and helpful comments. The work was supported by NSFC(60472017,90103033), MOE(KP0302) and MOST(ICPDF2003).",

year = "2005",

month = mar,

doi = "10.1016/j.neucom.2004.11.028",

language = "English",

volume = "64",

pages = "529--536",

journal = "Neurocomputing",

issn = "0925-2312",

number = "1-4 SPEC. ISS.",

}

TY - JOUR

T1 - A fast fixed-point algorithm for complexity pursuit

AU - Shi, Zhenwei

AU - Tang, Huanwen

AU - Tang, Yiyuan

N1 - Funding Information: We are grateful to all the anonymous reviewers who provided insightful and helpful comments. The work was supported by NSFC(60472017,90103033), MOE(KP0302) and MOST(ICPDF2003).

PY - 2005/3

Y1 - 2005/3

N2 - Complexity pursuit is a recently developed algorithm using the gradient descent for separating interesting components from time series. It is an extension of projection pursuit to time series data and the method is closely related to blind separation of time-dependent source signals and independent component analysis (ICA). In this paper, a fixed-point algorithm for complexity pursuit is introduced. The fixed-point algorithm inherits the advantages of the well-known FastICA algorithm in ICA, which is very simple, converges fast, and does not need choose any learning step sizes.

AB - Complexity pursuit is a recently developed algorithm using the gradient descent for separating interesting components from time series. It is an extension of projection pursuit to time series data and the method is closely related to blind separation of time-dependent source signals and independent component analysis (ICA). In this paper, a fixed-point algorithm for complexity pursuit is introduced. The fixed-point algorithm inherits the advantages of the well-known FastICA algorithm in ICA, which is very simple, converges fast, and does not need choose any learning step sizes.

KW - Blind source separation

KW - Complexity pursuit

KW - Independent component analysis

KW - Projection pursuit

KW - Time series

UR - http://www.scopus.com/inward/record.url?scp=15844402833&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2004.11.028

DO - 10.1016/j.neucom.2004.11.028

M3 - Article

AN - SCOPUS:15844402833

SN - 0925-2312

VL - 64

SP - 529

EP - 536

JO - Neurocomputing

JF - Neurocomputing

IS - 1-4 SPEC. ISS.

ER -

A fast fixed-point algorithm for complexity pursuit

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this