Abstract
This paper compares two Self-Organizing Map (SOM) based models for temporal sequence processing (TSP) both analytically and experimentally. These models, Temporal Kohonen Map (TKM) and Recurrent Self-Organizing Map (RSOM), incorporate leaky integrator memory to preserve the temporal context of the input signals. The learning and the convergence properties of the TKM and RSOM are studied and we show analytically that the RSOM is a significant improvement over the TKM, because the RSOM allows simple derivation of a consistent learning rule. The results of the analysis are demonstrated with experiments.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Bishop, C. M.: Neural Networks for Pattern Recognition, Oxford University Press, (1995).
Carpinteiro, O. A. S.: A hierarchical self-organizing map model for sequence recognition, In: Proceedings of ICANN '98, September (1998), pp. 816-820.
Chappel, G. J. and Taylor, J. G.: The temporal Kohenen map, Neural Networks, 6 (1993), 441-445
Cottrell, M.: Theoretical aspects of the som algorithm, In: Proceedings of WSOM '97, Workshop on Self-Organizing Maps, Helsinki University of Technology, Neural Networks Research Centre, Espoo, Finland, June (1997), pp. 246-267.
Cottrell, M.: Theoretical aspects of the som algorithm, Neurocomputing, 21 (1998), 119-138.
Erwin, E., Obermeyer, K. and Schulten, K.: Self-organizing maps: ordering, convergence properties and energy functions, Biological Cybenetics, 67 (1992), 47-55.
Kangas, J.: On the Analysis of Pattern Sequences by Self-OrganizingMaps, PhD thesis, Helsinki University of Technology, Espoo, Finland, (1994).
Kohenen, T.: The hypermap architecture, In: T. Kohenen, K. MÌkisara, O. Simula and J. Kangas, (eds), Artificial Neural Networks, Vol. II, Amsterdam, Netherlands, North-Holland, (1991), pp. 1357-1360.
Kohenen, T.: Things you have'nt heard about the Self-Organizing Map, In: Proceedings of the ICNN '93, International Conference on Neural Networks, Piscataway, NJ, IEEE, IEEE Service Center, (1993), pp. 1147-1156.
Kohenen, T.: Self-Organizing Maps, Vol. 30 of Lecture Notes in Information Sciences, Springer, 2nd Edn, (1997).
Koskela, T., Varsta, M., Heikkonen, J. and Kaski, K.: Prediction using rsom with local linear models, Int Journal of Knowledge-Based Intelligent Engineering Systems, 2(1) (1998), 60-68.
Leinonen, L., Kangas, J., Torkkola, K. and Juvas, A.: Dysphonia detected by pattern recognition of spectral composition, J. Speech and Hearing Res., 35 April (1992), 287-295.
Luttrell, S. P.: Image compression using a multilayer neural network, Pattern Recognition Letters, 10 (1989), 1-7.
Proakis, G. and Manolakis, D. G.: Digital Signal Processing: Principles, Algorithms, and Applications, Macmillan Publishing Company, (1992).
Tryba, V. and Goser, K.: Self-Organizing Feature Maps for process control in chemistry. In: T. Kohonen, K. MÌkisara, O. Simula and J. Kangas, (eds), Artificial Neural Networks, Amsterdam, Netherlands, North-Holland, (1991), pp. 847-852.
Varsta, M., del Ruiz Millän, J. and Heikkonen, J.: A recurrent self-organizing map for temporal sequence processing, In: Proceedings of the ICANN '97, Springer-Verlag, Berlin, Heidelberg, New York, October 1997. ISBN 3-540-63631-5.
Varsta, M., Heikkonen, J. and del Ruiz Millän, J.: Context learning with the self organizing map, In: Proceedings of WSOM '97, Workshop on Self-Organizing Maps, Helsinki University of Technology, Neural Networks Research Centre, June 1997.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Varsta, M., Heikkonen, J., Lampinen, J. et al. Temporal Kohonen Map and the Recurrent Self-Organizing Map: Analytical and Experimental Comparison. Neural Processing Letters 13, 237–251 (2001). https://doi.org/10.1023/A:1011353011837
Issue Date:
DOI: https://doi.org/10.1023/A:1011353011837