Memory efficient low-rank non-linear subspace tracking

F. Sheikholeslami; D. Berberidis; G. B. Giannakis

doi:10.1109/CAMSAP.2017.8313099

Memory efficient low-rank non-linear subspace tracking

F. Sheikholeslami, D. Berberidis, G. B. Giannakis

Electrical and Computer Engineering

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

1 Scopus citations

Abstract

The task of low-rank subspace tracking is of paramount importance for feature extraction over streaming data. Considering the broad range of applications in which the data fail to adhere to a linear model, the present work proposes a nonlinear subspace tracking algorithm. The proposed algorithm can effectively learn and track an evolving non-linear subspace in an online fashion. The notion of non-linearity is accommodated via exploitation of kernel-induced mappings, whose computational as well as memory requirements, if untreated, will impose scalability issues in large datasets. This issue is addressed by imposing a predefined affordable budget on the number of data vectors to be stored, preventing computational and memory growth of the algorithm, while enabling the tracking of possibly evolving subspaces. Numerical tests corroborate the effectiveness of the proposed algorithm on synthetic as well as real datasets.

Original language	English (US)
Title of host publication	2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1-5
Number of pages	5
ISBN (Electronic)	9781538612514
DOIs	https://doi.org/10.1109/CAMSAP.2017.8313099
State	Published - Mar 9 2018
Event	7th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017 - Curacao Duration: Dec 10 2017 → Dec 13 2017

Publication series

Name	2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017
Volume	2017-December

Conference

Conference	7th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017
City	Curacao
Period	12/10/17 → 12/13/17

Bibliographical note

Funding Information:
∗This work was supported by NSF grants 1500713, 1514056, and the NIH grant no. 1R01GM104975-01. E-mails: {sheik081,bermp001,georgios}@umn.edu

Publisher Copyright:
© 2017 IEEE.

Keywords

Nonlinear subspace tracking
budgeted learning
kernel methods
online learning

Access

10.1109/CAMSAP.2017.8313099

OpenUrl availability

Full text

Cite this

Sheikholeslami, F., Berberidis, D., & Giannakis, G. B. (2018). Memory efficient low-rank non-linear subspace tracking. In 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017 (pp. 1-5). Article 8313099 (2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017; Vol. 2017-December). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CAMSAP.2017.8313099

Memory efficient low-rank non-linear subspace tracking. / Sheikholeslami, F.; Berberidis, D.; Giannakis, G. B.
2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017. Institute of Electrical and Electronics Engineers Inc., 2018. p. 1-5 8313099 (2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017; Vol. 2017-December).

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

Sheikholeslami, F, Berberidis, D & Giannakis, GB 2018, Memory efficient low-rank non-linear subspace tracking. in 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017., 8313099, 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017, vol. 2017-December, Institute of Electrical and Electronics Engineers Inc., pp. 1-5, 7th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017, Curacao, 12/10/17. https://doi.org/10.1109/CAMSAP.2017.8313099

Sheikholeslami F, Berberidis D, Giannakis GB. Memory efficient low-rank non-linear subspace tracking. In 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017. Institute of Electrical and Electronics Engineers Inc. 2018. p. 1-5. 8313099. (2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017). doi: 10.1109/CAMSAP.2017.8313099

Sheikholeslami, F. ; Berberidis, D. ; Giannakis, G. B. / Memory efficient low-rank non-linear subspace tracking. 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 1-5 (2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017).

@inproceedings{1aa369850b954d02b302380fadd84e77,

title = "Memory efficient low-rank non-linear subspace tracking",

abstract = "The task of low-rank subspace tracking is of paramount importance for feature extraction over streaming data. Considering the broad range of applications in which the data fail to adhere to a linear model, the present work proposes a nonlinear subspace tracking algorithm. The proposed algorithm can effectively learn and track an evolving non-linear subspace in an online fashion. The notion of non-linearity is accommodated via exploitation of kernel-induced mappings, whose computational as well as memory requirements, if untreated, will impose scalability issues in large datasets. This issue is addressed by imposing a predefined affordable budget on the number of data vectors to be stored, preventing computational and memory growth of the algorithm, while enabling the tracking of possibly evolving subspaces. Numerical tests corroborate the effectiveness of the proposed algorithm on synthetic as well as real datasets.",

keywords = "Nonlinear subspace tracking, budgeted learning, kernel methods, online learning",

author = "F. Sheikholeslami and D. Berberidis and Giannakis, {G. B.}",

note = "Funding Information: ∗This work was supported by NSF grants 1500713, 1514056, and the NIH grant no. 1R01GM104975-01. E-mails: {sheik081,bermp001,georgios}@umn.edu Publisher Copyright: {\textcopyright} 2017 IEEE.; 7th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017 ; Conference date: 10-12-2017 Through 13-12-2017",

year = "2018",

month = mar,

day = "9",

doi = "10.1109/CAMSAP.2017.8313099",

language = "English (US)",

series = "2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1--5",

booktitle = "2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017",

}

TY - GEN

T1 - Memory efficient low-rank non-linear subspace tracking

AU - Sheikholeslami, F.

AU - Berberidis, D.

AU - Giannakis, G. B.

N1 - Funding Information: ∗This work was supported by NSF grants 1500713, 1514056, and the NIH grant no. 1R01GM104975-01. E-mails: {sheik081,bermp001,georgios}@umn.edu Publisher Copyright: © 2017 IEEE.

PY - 2018/3/9

Y1 - 2018/3/9

N2 - The task of low-rank subspace tracking is of paramount importance for feature extraction over streaming data. Considering the broad range of applications in which the data fail to adhere to a linear model, the present work proposes a nonlinear subspace tracking algorithm. The proposed algorithm can effectively learn and track an evolving non-linear subspace in an online fashion. The notion of non-linearity is accommodated via exploitation of kernel-induced mappings, whose computational as well as memory requirements, if untreated, will impose scalability issues in large datasets. This issue is addressed by imposing a predefined affordable budget on the number of data vectors to be stored, preventing computational and memory growth of the algorithm, while enabling the tracking of possibly evolving subspaces. Numerical tests corroborate the effectiveness of the proposed algorithm on synthetic as well as real datasets.

AB - The task of low-rank subspace tracking is of paramount importance for feature extraction over streaming data. Considering the broad range of applications in which the data fail to adhere to a linear model, the present work proposes a nonlinear subspace tracking algorithm. The proposed algorithm can effectively learn and track an evolving non-linear subspace in an online fashion. The notion of non-linearity is accommodated via exploitation of kernel-induced mappings, whose computational as well as memory requirements, if untreated, will impose scalability issues in large datasets. This issue is addressed by imposing a predefined affordable budget on the number of data vectors to be stored, preventing computational and memory growth of the algorithm, while enabling the tracking of possibly evolving subspaces. Numerical tests corroborate the effectiveness of the proposed algorithm on synthetic as well as real datasets.

KW - Nonlinear subspace tracking

KW - budgeted learning

KW - kernel methods

KW - online learning

UR - http://www.scopus.com/inward/record.url?scp=85043289844&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043289844&partnerID=8YFLogxK

U2 - 10.1109/CAMSAP.2017.8313099

DO - 10.1109/CAMSAP.2017.8313099

M3 - Conference contribution

AN - SCOPUS:85043289844

T3 - 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017

SP - 1

EP - 5

BT - 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 7th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2017

Y2 - 10 December 2017 through 13 December 2017

ER -

Memory efficient low-rank non-linear subspace tracking

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this