Learning complex-valued latent filters with absolute cosine similarity

Anh H.T. Nguyen; V. G. Reju; Andy W.H. Khong; Ing Yann Soon

doi:10.1109/ICASSP.2017.7952589

Learning complex-valued latent filters with absolute cosine similarity

Anh H.T. Nguyen^*, V. G. Reju, Andy W.H. Khong, Ing Yann Soon

^*Corresponding author for this work

Nanyang Technological University

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

6 Citations (Scopus)

Abstract

We propose a new sparse coding technique based on the power mean of phase-invariant cosine distances. Our approach is a generalization of sparse filtering and K-hyperlines clustering. It offers a better sparsity enforcer than the L₁/L₂ norm ratio that is typically used in sparse filtering. At the same time, the proposed approach scales better than the clustering counterparts for high-dimensional input. Our algorithm fully exploits the prior information obtained by preprocessing the observed data with whitening via an efficient row-wise decoupling scheme. In our simulating experiments, the algorithm produces better estimates than previous approaches do. It yields better separation of live recorded speech mixtures as well.

Original language	English
Title of host publication	2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	2412-2416
Number of pages	5
ISBN (Electronic)	9781509041176
DOIs	https://doi.org/10.1109/ICASSP.2017.7952589
Publication status	Published - Jun 16 2017
Externally published	Yes
Event	2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - New Orleans, United States Duration: Mar 5 2017 → Mar 9 2017

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)	1520-6149

Conference

Conference	2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017
Country/Territory	United States
City	New Orleans
Period	3/5/17 → 3/9/17

Bibliographical note

Publisher Copyright:
© 2017 IEEE.

ASJC Scopus Subject Areas

Software
Signal Processing
Electrical and Electronic Engineering

Keywords

blind source separation
cosine similarity
Row-wise decoupling
sparse component analysis

Access to Document

10.1109/ICASSP.2017.7952589

Cite this

Nguyen, A. H. T., Reju, V. G., Khong, A. W. H., & Soon, I. Y. (2017). Learning complex-valued latent filters with absolute cosine similarity. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings (pp. 2412-2416). Article 7952589 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICASSP.2017.7952589

Nguyen, Anh H.T. ; Reju, V. G. ; Khong, Andy W.H. et al. / Learning complex-valued latent filters with absolute cosine similarity. 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 2412-2416 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{ad6606d6d01d4e1b80606968ee1515a9,

title = "Learning complex-valued latent filters with absolute cosine similarity",

abstract = "We propose a new sparse coding technique based on the power mean of phase-invariant cosine distances. Our approach is a generalization of sparse filtering and K-hyperlines clustering. It offers a better sparsity enforcer than the L1/L2 norm ratio that is typically used in sparse filtering. At the same time, the proposed approach scales better than the clustering counterparts for high-dimensional input. Our algorithm fully exploits the prior information obtained by preprocessing the observed data with whitening via an efficient row-wise decoupling scheme. In our simulating experiments, the algorithm produces better estimates than previous approaches do. It yields better separation of live recorded speech mixtures as well.",

keywords = "blind source separation, cosine similarity, Row-wise decoupling, sparse component analysis",

author = "Nguyen, {Anh H.T.} and Reju, {V. G.} and Khong, {Andy W.H.} and Soon, {Ing Yann}",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 ; Conference date: 05-03-2017 Through 09-03-2017",

year = "2017",

month = jun,

day = "16",

doi = "10.1109/ICASSP.2017.7952589",

language = "English",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

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

pages = "2412--2416",

booktitle = "2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings",

address = "United States",

}

Nguyen, AHT, Reju, VG, Khong, AWH & Soon, IY 2017, Learning complex-valued latent filters with absolute cosine similarity. in 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings., 7952589, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 2412-2416, 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017, New Orleans, United States, 3/5/17. https://doi.org/10.1109/ICASSP.2017.7952589

Learning complex-valued latent filters with absolute cosine similarity. / Nguyen, Anh H.T.; Reju, V. G.; Khong, Andy W.H. et al.
2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2017. p. 2412-2416 7952589 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

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

TY - GEN

T1 - Learning complex-valued latent filters with absolute cosine similarity

AU - Nguyen, Anh H.T.

AU - Reju, V. G.

AU - Khong, Andy W.H.

AU - Soon, Ing Yann

PY - 2017/6/16

Y1 - 2017/6/16

N2 - We propose a new sparse coding technique based on the power mean of phase-invariant cosine distances. Our approach is a generalization of sparse filtering and K-hyperlines clustering. It offers a better sparsity enforcer than the L1/L2 norm ratio that is typically used in sparse filtering. At the same time, the proposed approach scales better than the clustering counterparts for high-dimensional input. Our algorithm fully exploits the prior information obtained by preprocessing the observed data with whitening via an efficient row-wise decoupling scheme. In our simulating experiments, the algorithm produces better estimates than previous approaches do. It yields better separation of live recorded speech mixtures as well.

AB - We propose a new sparse coding technique based on the power mean of phase-invariant cosine distances. Our approach is a generalization of sparse filtering and K-hyperlines clustering. It offers a better sparsity enforcer than the L1/L2 norm ratio that is typically used in sparse filtering. At the same time, the proposed approach scales better than the clustering counterparts for high-dimensional input. Our algorithm fully exploits the prior information obtained by preprocessing the observed data with whitening via an efficient row-wise decoupling scheme. In our simulating experiments, the algorithm produces better estimates than previous approaches do. It yields better separation of live recorded speech mixtures as well.

KW - blind source separation

KW - cosine similarity

KW - Row-wise decoupling

KW - sparse component analysis

UR - http://www.scopus.com/inward/record.url?scp=85023741737&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85023741737&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2017.7952589

DO - 10.1109/ICASSP.2017.7952589

M3 - Conference contribution

AN - SCOPUS:85023741737

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 2412

EP - 2416

BT - 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017

Y2 - 5 March 2017 through 9 March 2017

ER -

Nguyen AHT, Reju VG, Khong AWH, Soon IY. Learning complex-valued latent filters with absolute cosine similarity. In 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2017. p. 2412-2416. 7952589. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP.2017.7952589

Learning complex-valued latent filters with absolute cosine similarity

Abstract

Publication series

Conference

Bibliographical note

ASJC Scopus Subject Areas

Keywords

Access to Document

Other files and links

Fingerprint

Cite this