Publication

Learning vector quantization and relevances in complex coefficient space

Straat, M., Kaden, M., Gay, M., Villmann, T., Lampe, A., Seiffert, U., Biehl, M. & Melchert, F., 9-Mar-2019, In : Neural Computing and Applications. 15 p.

Research output: Contribution to journalArticleAcademicpeer-review

APA

Straat, M., Kaden, M., Gay, M., Villmann, T., Lampe, A., Seiffert, U., ... Melchert, F. (2019). Learning vector quantization and relevances in complex coefficient space. Neural Computing and Applications. https://doi.org/10.1007/s00521-019-04080-5

Author

Straat, M. ; Kaden, M. ; Gay, M. ; Villmann, T. ; Lampe, A. ; Seiffert, U. ; Biehl, M. ; Melchert, F. / Learning vector quantization and relevances in complex coefficient space. In: Neural Computing and Applications. 2019.

Harvard

Straat, M, Kaden, M, Gay, M, Villmann, T, Lampe, A, Seiffert, U, Biehl, M & Melchert, F 2019, 'Learning vector quantization and relevances in complex coefficient space', Neural Computing and Applications. https://doi.org/10.1007/s00521-019-04080-5

Standard

Learning vector quantization and relevances in complex coefficient space. / Straat, M.; Kaden, M.; Gay, M.; Villmann, T.; Lampe, A.; Seiffert, U.; Biehl, M.; Melchert, F.

In: Neural Computing and Applications, 09.03.2019.

Research output: Contribution to journalArticleAcademicpeer-review

Vancouver

Straat M, Kaden M, Gay M, Villmann T, Lampe A, Seiffert U et al. Learning vector quantization and relevances in complex coefficient space. Neural Computing and Applications. 2019 Mar 9. https://doi.org/10.1007/s00521-019-04080-5


BibTeX

@article{0328534594c544458637d750b79874f0,
title = "Learning vector quantization and relevances in complex coefficient space",
abstract = "In this contribution, we consider the classification of time series and similar functional data which can be represented in complex Fourier and wavelet coefficient space. We apply versions of learning vector quantization (LVQ) which are suitable for complex-valued data, based on the so-called Wirtinger calculus. It allows for the formulation of gradient-based update rules in the framework of cost-function-based generalized matrix relevance LVQ (GMLVQ). Alternatively, we consider the concatenation of real and imaginary parts of Fourier coefficients in a real-valued feature vector and the classification of time-domain representations by means of conventional GMLVQ. In addition, we consider the application of the method in combination with wavelet-space features to heartbeat classification.",
author = "M. Straat and M. Kaden and M. Gay and T. Villmann and A. Lampe and U. Seiffert and M. Biehl and F. Melchert",
year = "2019",
month = "3",
day = "9",
doi = "10.1007/s00521-019-04080-5",
language = "English",
journal = "Neural Computing and Applications",
issn = "1433-3058",
publisher = "SPRINGER LONDON LTD",

}

RIS

TY - JOUR

T1 - Learning vector quantization and relevances in complex coefficient space

AU - Straat, M.

AU - Kaden, M.

AU - Gay, M.

AU - Villmann, T.

AU - Lampe, A.

AU - Seiffert, U.

AU - Biehl, M.

AU - Melchert, F.

PY - 2019/3/9

Y1 - 2019/3/9

N2 - In this contribution, we consider the classification of time series and similar functional data which can be represented in complex Fourier and wavelet coefficient space. We apply versions of learning vector quantization (LVQ) which are suitable for complex-valued data, based on the so-called Wirtinger calculus. It allows for the formulation of gradient-based update rules in the framework of cost-function-based generalized matrix relevance LVQ (GMLVQ). Alternatively, we consider the concatenation of real and imaginary parts of Fourier coefficients in a real-valued feature vector and the classification of time-domain representations by means of conventional GMLVQ. In addition, we consider the application of the method in combination with wavelet-space features to heartbeat classification.

AB - In this contribution, we consider the classification of time series and similar functional data which can be represented in complex Fourier and wavelet coefficient space. We apply versions of learning vector quantization (LVQ) which are suitable for complex-valued data, based on the so-called Wirtinger calculus. It allows for the formulation of gradient-based update rules in the framework of cost-function-based generalized matrix relevance LVQ (GMLVQ). Alternatively, we consider the concatenation of real and imaginary parts of Fourier coefficients in a real-valued feature vector and the classification of time-domain representations by means of conventional GMLVQ. In addition, we consider the application of the method in combination with wavelet-space features to heartbeat classification.

U2 - 10.1007/s00521-019-04080-5

DO - 10.1007/s00521-019-04080-5

M3 - Article

JO - Neural Computing and Applications

JF - Neural Computing and Applications

SN - 1433-3058

ER -

ID: 77709390