Abstract
Neural Gas is a neural network algorithm for vector quantization. It has not arbitrary established network topology, instead its topology is changing dynamically during training process. Originally, the Neural Gas is an unsupervised algorithm. However, there are several extensions that enables Neural Gas to use the information about sample’s class. This significantly improves the accuracy of obtained clusters. Therefore, the Neural Gas was successfully used in classification problems. In this paper we present a novel method to learn the Neural Gas with fully and partially labelled data sets. Proposed method simulates the neuron’s hesitation between membership to the classes during the learning. Hesitation process is based on neuron’s class membership probability and Metropolis-Hastings algorithm. The proposed method was compared with state-of-art extensions of Neural Gas on supervised and semi-supervised classification tasks on benchmark data sets. Experimental results yield better or the same classification accuracy on both types of supervision.
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.
Similar content being viewed by others
References
Asuncion, A., Newman, D.J.: UCI machine learning repository. University of California, Irvine, School of Information and Computer Sciences (2007)
Du, K.-L.: Clustering: A neural network approach. Neural Networks 23, 89–107 (2010)
Fritzke, B.: A Growing Neural Gas Network Learns Topologies. In: Advances in Neural Information Processing Systems (NIPS 1994), pp. 625–632 (1994)
Hammer, B., Hasenfuss, A., Schleif, F.-M., Villmann, T.: Supervised Batch Neural Gas. In: Schwenker, F., Marinai, S. (eds.) ANNPR 2006. LNCS (LNAI), vol. 4087, pp. 33–45. Springer, Heidelberg (2006)
Hammer, B., Strickert, M., Villmann, T.: Supervised Neural Gas with General Similarity Measure. Neural Processing Letters 21, 21–44 (2005)
Hastings, W.K.: Monte Carlo Sampling Methods Using Markov Chains and Their Applications. Biometrika 57, 97–109 (1970)
Herrmann, M., Villmann, T.: Vector Quantization by Optimal Neural Gas. In: Gerstner, W., Hasler, M., Germond, A., Nicoud, J.-D. (eds.) ICANN 1997. LNCS, vol. 1327, pp. 625–630. Springer, Heidelberg (1997)
Kästner, M., Villmann, T.: Fuzzy Supervised Self-Organizing Map for Semi-supervised Vector Quantization. In: Rutkowski, L., Korytkowski, M., Scherer, R., Tadeusiewicz, R., Zadeh, L.A., Zurada, J.M. (eds.) ICAISC 2012, Part I. LNCS, vol. 7267, pp. 256–265. Springer, Heidelberg (2012)
Kirkpatrick, S., Gelatt, C.D., Vecchi, M.P.: Optimization by Simulated Annealing. Science 220, 671–680 (1983)
Kohonen, T.: The Self-Organizing Map. Proceedings of the IEEE 78, 1464–1480 (1990)
Martinetz, T., Schulten, K.: A Neural-Gas Network Learns Topologies. Artificial Neural Networks 1, 397–402 (1991)
Metropolis, N., Rosenbluth, A.W., Rosenbluth, M.N., Teller, A.H., Teller, E.: Equations of State Calculations by Fast Computing Machines. Journal of Chemical Physics 21, 1087–1092 (1953)
Midenet, S., Grumbach, A.: Learning Associations by Self-Organization: The LASSO model. Neurocomputing 6, 343–361 (1994)
Möller, R., Hoffmann, H.: An extension of neural gas to local PCA. Neurocomputing 62, 305–326 (2004)
Płoński, P., Zaremba, K.: Self-Organising Maps for Classification with Metropolis-Hastings Algorithm for Supervision. In: Huang, T., Zeng, Z., Li, C., Leung, C.S. (eds.) ICONIP 2012, Part III. LNCS, vol. 7665, pp. 149–156. Springer, Heidelberg (2012)
Schenck, W., Welsch, R., Kaiser, A., Möller, R.: Adaptive learning rate control for neural gas principal component analysis. In: European Symposium on Artificial Neural Networks (ESANN 2010), pp. 213–218. d-side pub. (2010)
Schleif, F.-M., Villmann, T., Hammer, B.: Supervised Neural Gas for Classification of Functional Data and Its Application to the Analysis of Clinical Proteom Spectra. In: Sandoval, F., Prieto, A.G., Cabestany, J., Graña, M. (eds.) IWANN 2007. LNCS, vol. 4507, pp. 1036–1044. Springer, Heidelberg (2007)
Villmann, T., Geweniger, T., Kästner, M., Lange, M.: Fuzzy Neural Gas for Unsupervised Vector Quantization. In: Rutkowski, L., Korytkowski, M., Scherer, R., Tadeusiewicz, R., Zadeh, L.A., Zurada, J.M. (eds.) ICAISC 2012, Part I. LNCS, vol. 7267, pp. 350–358. Springer, Heidelberg (2012)
Villmann, T., Hammer, B., Schleif, F.-M., Geweniger, T., Hermann, W.: Fuzzy classification by fuzzy labeled neural gas. Neural Networks 19, 772–779 (2006)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Płoński, P., Zaremba, K. (2013). Hesitant Neural Gas for Supervised and Semi-supervised Classification. In: Rutkowski, L., Korytkowski, M., Scherer, R., Tadeusiewicz, R., Zadeh, L.A., Zurada, J.M. (eds) Artificial Intelligence and Soft Computing. ICAISC 2013. Lecture Notes in Computer Science(), vol 7894. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-38658-9_42
Download citation
DOI: https://doi.org/10.1007/978-3-642-38658-9_42
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-38657-2
Online ISBN: 978-3-642-38658-9
eBook Packages: Computer ScienceComputer Science (R0)