Abstract
The general aim in classification learning by supervised training is to achieve a high classification performance, frequently judged in terms of classification accuracy. A powerful method is the generalized learning vector quantizer, which realizes a gradient based optimization scheme based on a cost function approximating the usual symmetric misclassification rate. In this paper we investigate a modification of this approach taking into account asymmetric misclassification penalties to reflect structural knowledge of external experts about the data, as it is frequently the case for instance in medicine. Further we also discuss the weighting of importance for the considered classes in the classification problem. We show that both aspects can be seen as a kind of attention based learning strategy.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Barthel, H., Villmann, T., Hermann, W., Hesse, S., Kühn, H.-J., Wagner, A., Kluge, R.: Different patterns of brain glucose consumption in Wilsons disease. Zeitschrift für Gastroenterologie 39, 241 (2001)
Biehl, M., Kästner, M., Lange, M., Villmann, T.: Non-Euclidean principal component analysis and Oja’s learning rule – theoretical aspects. In: Estevez, P.A., Principe, J.C., Zegers, P. (eds.) Advances in Self-Organizing Maps. AISC, vol. 198, pp. 23–34. Springer, Heidelberg (2013)
Der, R., Herrmann, M.: Attention based partitioning. In: der Meer, M.V. (ed.) Bericht Des Status–Seminar Des BMFT Neuroinformatik, pp. 441–446. DLR, Berlin (1992)
Duda, R., Hart, P.: Pattern Classification and Scene Analysis. Wiley, NY (1973)
Hammer, B., Villmann, T.: Generalized relevance learning vector quantization. Neural Networks 15(8-9), 1059–1068 (2002)
Hermann, W., Caca, K., Eggers, B., Villmann, T., Clark, D., Berr, F., Wagner, A.: Genotype correlation with fine motor symptoms in patients with Wilson’s disease. European Neurology 48, 97–101 (2002)
Hermann, W., Günther, P., Kühn, H.-J., Schneider, J., Eichelkraut, S., Villmann, T., Strecker, K., Schwarz, J., Wagner, A.: FAEP und Morphometrie des Mesenzephalons bei Morbus Wilson. Aktuelle Neurologie 34(10), 547–554 (2007)
Hermann, W., Villmann, T., Grahmann, F., Kühn, H., Wagner, A.: Investigation of fine motoric disturbances in Wilson’s disease. Neurological Sciences 23(6), 279–285 (2003)
Hermann, W., Villmann, T., Wagner, A.: Elektrophysiologisches Schädigungsprofil von Patienten mit einem Morbus Wilson’. Der Nervenarzt 74(10), 881–887 (2003)
Kaden, M., Hermann, W., Villmann, T.: Optimization of general statistical accuracy measures for classification based on learning vector quantization. In: Verleysen, M. (ed.) Proc. of European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2014), Louvain-La-Neuve, Belgium (page accepted, 2014), i6doc.com
Kaden, M., Villmann, T.: A framework for optimization of statistical classification measures based on generalized learning vector quantization. Machine Learning Reports, 7(MLR-02-2013), 69–76 (2013), http://www.techfak.uni-bielefeld.de/~fschleif/mlr/mlr_02_2013.pdf , ISSN:1865-3960
Kästner, M., Riedel, M., Strickert, M., Hermann, W., Villmann, T.: Border-sensitive learning in kernelized learning vector quantization. In: Rojas, I., Joya, G., Gabestany, J. (eds.) IWANN 2013, Part I. LNCS, vol. 7902, pp. 357–366. Springer, Heidelberg (2013)
Lange, M., Biehl, M., Villmann, T.: Non-Euclidean principal component analysis by Hebbian learning. Neurocomputing (page in press, 2014)
Lange, M., Villmann, T.: Derivatives of l p-norms and their approximations. Machine Learning Reports 7(MLR-04-2013), 43–59 (2013), http://www.techfak.uni-bielefeld.de/~fschleif/mlr/mlr_04_013.pdf , ISSN:1865-3960
Matthews, B.: Comparison of the predicted and observed secondary structure of T4 phage Iysozyme. Biochimica et Biophysica Acta 405, 442–451 (1975)
Rijsbergen, C.: Information Retrieval, 2nd edn. Butterworths, London (1979)
Sachs, L.: Angewandte Statistik, 7th edn. Springer (1992)
Sato, A.S., Yamada, K.: Generalized learning vector quantization. In: Tesauro, G., Touretzky, D., Leen, T. (eds.) Advances in Neural Information Processing Systems, vol. 7, pp. 423–429. MIT Press (1995)
Schneider, P., Hammer, B., Biehl, M.: Adaptive relevance matrices in learning vector quantization. Neural Computation 21, 3532–3561 (2009)
Strickert, M., Schleif, F.-M., Seiffert, U., Villmann, T.: Derivatives of Pearson correlation for gradient-based analysis of biomedical data. Inteligencia Artificial, Revista Iberoamericana de Inteligencia Artificial (37), 37–44 (2008)
Villmann, T., Haase, S.: Divergence based vector quantization. Neural Computation 23(5), 1343–1392 (2011)
Villmann, T., Haase, S., Kaden, M.: Kernelized vector quantization in gradient-descent learning. Neurocomputing (page in press, 2014)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Kaden, M., Hermann, W., Villmann, T. (2014). Attention Based Classification Learning in GLVQ and Asymmetric Misclassification Assessment. In: Villmann, T., Schleif, FM., Kaden, M., Lange, M. (eds) Advances in Self-Organizing Maps and Learning Vector Quantization. Advances in Intelligent Systems and Computing, vol 295. Springer, Cham. https://doi.org/10.1007/978-3-319-07695-9_7
Download citation
DOI: https://doi.org/10.1007/978-3-319-07695-9_7
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-07694-2
Online ISBN: 978-3-319-07695-9
eBook Packages: EngineeringEngineering (R0)