ICANN 1996: Artificial Neural Networks — ICANN 96 pp 821-826 | Cite as
Building nonlinear data models with self-organizing maps
Poster Presentations 3 Theory V: Self-Organizing Maps
First Online:
Abstract
We study the extraction of nonlinear data models in high dimensional spaces with modified self-organizing maps. Our algorithm maps lower dimensional lattice into a high dimensional space without topology violations by tuning the neighborhood widths locally. The approach is based on a new principle exploiting the specific dynamical properties of the first order phase transition induced by the noise of the data. The performance of the algorithm is demonstrated for one- and two-dimensional principal manifolds and for sparse data sets.
Keywords
High Dimensional Space Neighborhood Parameter Neighborhood Width Principal Curf Principal Manifold
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
Preview
Unable to display preview. Download preview PDF.
References
- 1.T. Hastie, W. Stuetzle, Principal curves. Journal of the American Statistical Association 84, 502–516 (1989)Google Scholar
- 2.T. Kohonen, The Self-Organizing Map, Springer-Verlag, 1995.Google Scholar
- 3.H. Ritter, T. Martinetz, K. Schulten, Neural Computation and Self-Organizing Maps, Addison-Wesley, 1992.Google Scholar
- 4.F. Mulier, V. Cherkassky, Self-Organization as an Iterative Kernel Smoothing Process. Neural Computation 7, 1165–1177, 1995.Google Scholar
- 5.R. Der, M. Herrmann, Critical Phenomena in Self-Organized Feature Maps: A Ginzburg-Landau Approach, Phys. Rev. E 49:6, 5840–5848, 1994.Google Scholar
- 6.M. Herrmann, “Self-Organizing Feature Maps with Self-Organizing Neighborhood Widths”, Proc. 1995 IEEE Intern. Conf. on Neur. Networks, 2998–3003, 1995.Google Scholar
- 7.C. K. Chui, An Introduction to Wavelets, Academic Press, 1992.Google Scholar
Copyright information
© Springer-Verlag Berlin Heidelberg 1996