Generalized Derivative Based Kernelized Learning Vector Quantization

Frank-Michael Schleif; Thomas Villmann; Barbara Hammer; Petra Schneider; Michael Biehl

doi:10.1007/978-3-642-15381-5_3

Generalized Derivative Based Kernelized Learning Vector Quantization

Frank-Michael Schleif, Thomas Villmann, Barbara Hammer, Petra Schneider, Michael Biehl

Intelligent Systems

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

11 Citations (Scopus)

290 Downloads (Pure)

Abstract

We derive a novel derivative based version of kernelized Generalized Learning Vector Quantization (KGLVQ) as an effective, easy to interpret, prototype based and kernelized classifier. It is called D-KGLVQ and we provide generalization error bounds, experimental results on real world data, showing that D-KGLVQ is competitive with KGLVQ and the SVM on UCI data and additionally show that automatic parameter adaptation for the used kernels simplifies the learning.

Original language	English
Title of host publication	Proc. Intelligent Data Engineering and Automated Learning - IDEAL 2010
Editors	Colin Fyfe, Peter Tino, Darryl Charles, Cesar Garcia-Osoro, Hujun Yin
Publisher	Springer
Pages	21-28
Number of pages	8
DOIs	https://doi.org/10.1007/978-3-642-15381-5_3
Publication status	Published - 2010

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	6283

Access to Document

10.1007/978-3-642-15381-5_3

2010LNCSSchleif.pdfFinal publisher's version, 177 KB

Handle.net

Persistent link

Cite this

Schleif, F-M., Villmann, T., Hammer, B., Schneider, P., & Biehl, M. (2010). Generalized Derivative Based Kernelized Learning Vector Quantization. In C. Fyfe, P. Tino, D. Charles, C. Garcia-Osoro, & H. Yin (Eds.), Proc. Intelligent Data Engineering and Automated Learning - IDEAL 2010 (pp. 21-28). (Lecture Notes in Computer Science; Vol. 6283). Springer. https://doi.org/10.1007/978-3-642-15381-5_3

@inproceedings{239369d4fd494ac386bb969994397a9d,

title = "Generalized Derivative Based Kernelized Learning Vector Quantization",

abstract = "We derive a novel derivative based version of kernelized Generalized Learning Vector Quantization (KGLVQ) as an effective, easy to interpret, prototype based and kernelized classifier. It is called D-KGLVQ and we provide generalization error bounds, experimental results on real world data, showing that D-KGLVQ is competitive with KGLVQ and the SVM on UCI data and additionally show that automatic parameter adaptation for the used kernels simplifies the learning.",

author = "Frank-Michael Schleif and Thomas Villmann and Barbara Hammer and Petra Schneider and Michael Biehl",

note = "Relation: http://www.rug.nl/informatica/onderzoek/bernoulli Rights: University of Groningen, Johann Bernoulli Institute for Mathematics and Computer Science",

year = "2010",

doi = "10.1007/978-3-642-15381-5_3",

language = "English",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "21--28",

editor = "Colin Fyfe and Peter Tino and Darryl Charles and Cesar Garcia-Osoro and Hujun Yin",

booktitle = "Proc. Intelligent Data Engineering and Automated Learning - IDEAL 2010",

}

Schleif, F-M, Villmann, T, Hammer, B, Schneider, P & Biehl, M 2010, Generalized Derivative Based Kernelized Learning Vector Quantization. in C Fyfe, P Tino, D Charles, C Garcia-Osoro & H Yin (eds), Proc. Intelligent Data Engineering and Automated Learning - IDEAL 2010. Lecture Notes in Computer Science, vol. 6283, Springer, pp. 21-28. https://doi.org/10.1007/978-3-642-15381-5_3

Generalized Derivative Based Kernelized Learning Vector Quantization. / Schleif, Frank-Michael; Villmann, Thomas; Hammer, Barbara et al.
Proc. Intelligent Data Engineering and Automated Learning - IDEAL 2010. ed. / Colin Fyfe; Peter Tino; Darryl Charles; Cesar Garcia-Osoro; Hujun Yin. Springer, 2010. p. 21-28 (Lecture Notes in Computer Science; Vol. 6283).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Generalized Derivative Based Kernelized Learning Vector Quantization

AU - Schleif, Frank-Michael

AU - Villmann, Thomas

AU - Hammer, Barbara

AU - Schneider, Petra

AU - Biehl, Michael

N1 - Relation: http://www.rug.nl/informatica/onderzoek/bernoulli Rights: University of Groningen, Johann Bernoulli Institute for Mathematics and Computer Science

PY - 2010

Y1 - 2010

N2 - We derive a novel derivative based version of kernelized Generalized Learning Vector Quantization (KGLVQ) as an effective, easy to interpret, prototype based and kernelized classifier. It is called D-KGLVQ and we provide generalization error bounds, experimental results on real world data, showing that D-KGLVQ is competitive with KGLVQ and the SVM on UCI data and additionally show that automatic parameter adaptation for the used kernels simplifies the learning.

AB - We derive a novel derivative based version of kernelized Generalized Learning Vector Quantization (KGLVQ) as an effective, easy to interpret, prototype based and kernelized classifier. It is called D-KGLVQ and we provide generalization error bounds, experimental results on real world data, showing that D-KGLVQ is competitive with KGLVQ and the SVM on UCI data and additionally show that automatic parameter adaptation for the used kernels simplifies the learning.

U2 - 10.1007/978-3-642-15381-5_3

DO - 10.1007/978-3-642-15381-5_3

M3 - Conference contribution

T3 - Lecture Notes in Computer Science

SP - 21

EP - 28

BT - Proc. Intelligent Data Engineering and Automated Learning - IDEAL 2010

A2 - Fyfe, Colin

A2 - Tino, Peter

A2 - Charles, Darryl

A2 - Garcia-Osoro, Cesar

A2 - Yin, Hujun

PB - Springer

ER -