Abstract
Wave propagation within a “cortex” of neurons is used to influence the ordering of a self-organizing feature map. As a result, the network is able to represent temporal aspects of the input. Since the wave only uses local interactions between adjacent neurons, connectivity in the network is very low and a parallel hardware architecture suggests itself. Operation of a demonstration setup consisting of 16 neurons in digital technology is exemplified by the representation of phoneme sequences.
Preview
Unable to display preview. Download preview PDF.
References
T. Kohonen, Self-Organizing Maps (Springer, Berlin, 1995).
C. Darian-Smith, Ann. Rev. Psychol. 33, 155 (1982).
A. H. Cornell-Bell, S. M. Finkenbeiner, M. S. Cooper, S. J. Smith, Science 247, 470 (1990).
D. Ruwisch, M. Bode, H. G. Purwins, Neural Networks, 6, 1147 (1993).
D. Ruwisch, M. Bode, H.-J. Schulze, F. J. Niedernostheide, Lecture Notes in Physics 476, (Springer, Berlin, 1996).
J. Kangas, Proceedings of the international joint conference on neural networks (IJCNN) 2, 331–336 (1990).
M. C. Mozer, Predicting the future and understanding the past. (Addison Wesley, Redwood City, 1993).
D. L. James, R. Mükulainen, Advances in neural processing systems, 7 (1995).
N. R. Euliano, J. C. Principe, Proceedings of the international conference on neural networks (ICNN) 4, 1900–1905 (1996).
D. Ruwisch, B.Dobrzewski, and M. Bode, to appear in Proceeding of the 1997 IEEE Workshop on Neural Networks for Signal Processing (NNSP97), (1997).
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Dobrzewski, B., Ruwisch, D., Bode, M. (1997). Wave propagation in self-organizing feature maps as a means for the representation of temporal sequences. In: Gerstner, W., Germond, A., Hasler, M., Nicoud, JD. (eds) Artificial Neural Networks — ICANN'97. ICANN 1997. Lecture Notes in Computer Science, vol 1327. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0020230
Download citation
DOI: https://doi.org/10.1007/BFb0020230
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63631-1
Online ISBN: 978-3-540-69620-9
eBook Packages: Springer Book Archive