Information criteria for reduced rank canonical correlation analysis

Mohammed A. Hasan

doi:10.1109/IJCNN.2004.1380964

Information criteria for reduced rank canonical correlation analysis

Mohammed A. Hasan

Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Scopus citations

Abstract

Canonical correlation analysis is an essential technique in the field of multivariate statistical analysis. In this paper, a framework involving unconstrained optimization criteria is proposed for extracting multiple canonical variates and canonical correlations serially and in parallel. These criteria are derived from optimizing three information based functions. Based on the gradient-ascent or descent methods, we derive many algorithms for performing the true CCA recursively. The main feature of this approach is that orthogonal basis for canonical variates is automatically obtained. The first few singular values and vectors can also be obtained using this framework. The performance of the proposed algorithms is demonstrated through simulations.

Original language	English (US)
Title of host publication	2004 IEEE International Joint Conference on Neural Networks - Proceedings
Pages	2215-2220
Number of pages	6
DOIs	https://doi.org/10.1109/IJCNN.2004.1380964
State	Published - Dec 1 2004
Event	2004 IEEE International Joint Conference on Neural Networks - Proceedings - Budapest, Hungary Duration: Jul 25 2004 → Jul 29 2004

Publication series

Name	IEEE International Conference on Neural Networks - Conference Proceedings
Volume	3
ISSN (Print)	1098-7576

Other

Other	2004 IEEE International Joint Conference on Neural Networks - Proceedings
Country/Territory	Hungary
City	Budapest
Period	7/25/04 → 7/29/04

Keywords

Canonical correlation extraction
Fast adaptive algorithms
Information criteria

Publisher link

10.1109/IJCNN.2004.1380964

Find It

Find It at U of M Libraries

Cite this

Hasan, MA 2004, Information criteria for reduced rank canonical correlation analysis. in 2004 IEEE International Joint Conference on Neural Networks - Proceedings. IEEE International Conference on Neural Networks - Conference Proceedings, vol. 3, pp. 2215-2220, 2004 IEEE International Joint Conference on Neural Networks - Proceedings, Budapest, Hungary, 7/25/04. https://doi.org/10.1109/IJCNN.2004.1380964

@inproceedings{9af4a6f7b30d422ca956630057795808,

title = "Information criteria for reduced rank canonical correlation analysis",

abstract = "Canonical correlation analysis is an essential technique in the field of multivariate statistical analysis. In this paper, a framework involving unconstrained optimization criteria is proposed for extracting multiple canonical variates and canonical correlations serially and in parallel. These criteria are derived from optimizing three information based functions. Based on the gradient-ascent or descent methods, we derive many algorithms for performing the true CCA recursively. The main feature of this approach is that orthogonal basis for canonical variates is automatically obtained. The first few singular values and vectors can also be obtained using this framework. The performance of the proposed algorithms is demonstrated through simulations.",

keywords = "Canonical correlation extraction, Fast adaptive algorithms, Information criteria",

author = "Hasan, {Mohammed A.}",

year = "2004",

month = dec,

day = "1",

doi = "10.1109/IJCNN.2004.1380964",

language = "English (US)",

isbn = "0780383591",

series = "IEEE International Conference on Neural Networks - Conference Proceedings",

pages = "2215--2220",

booktitle = "2004 IEEE International Joint Conference on Neural Networks - Proceedings",

note = "2004 IEEE International Joint Conference on Neural Networks - Proceedings ; Conference date: 25-07-2004 Through 29-07-2004",

}

TY - GEN

T1 - Information criteria for reduced rank canonical correlation analysis

AU - Hasan, Mohammed A.

PY - 2004/12/1

Y1 - 2004/12/1

N2 - Canonical correlation analysis is an essential technique in the field of multivariate statistical analysis. In this paper, a framework involving unconstrained optimization criteria is proposed for extracting multiple canonical variates and canonical correlations serially and in parallel. These criteria are derived from optimizing three information based functions. Based on the gradient-ascent or descent methods, we derive many algorithms for performing the true CCA recursively. The main feature of this approach is that orthogonal basis for canonical variates is automatically obtained. The first few singular values and vectors can also be obtained using this framework. The performance of the proposed algorithms is demonstrated through simulations.

AB - Canonical correlation analysis is an essential technique in the field of multivariate statistical analysis. In this paper, a framework involving unconstrained optimization criteria is proposed for extracting multiple canonical variates and canonical correlations serially and in parallel. These criteria are derived from optimizing three information based functions. Based on the gradient-ascent or descent methods, we derive many algorithms for performing the true CCA recursively. The main feature of this approach is that orthogonal basis for canonical variates is automatically obtained. The first few singular values and vectors can also be obtained using this framework. The performance of the proposed algorithms is demonstrated through simulations.

KW - Canonical correlation extraction

KW - Fast adaptive algorithms

KW - Information criteria

UR - http://www.scopus.com/inward/record.url?scp=10844266528&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=10844266528&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.2004.1380964

DO - 10.1109/IJCNN.2004.1380964

M3 - Conference contribution

AN - SCOPUS:10844266528

SN - 0780383591

T3 - IEEE International Conference on Neural Networks - Conference Proceedings

SP - 2215

EP - 2220

BT - 2004 IEEE International Joint Conference on Neural Networks - Proceedings

T2 - 2004 IEEE International Joint Conference on Neural Networks - Proceedings

Y2 - 25 July 2004 through 29 July 2004

ER -

Information criteria for reduced rank canonical correlation analysis

Abstract

Publication series

Other

Keywords

Publisher link

Find It

Other files and links

Fingerprint

Cite this