Abstract
Kohonen and others have devised network algorithms for computing so-called topological feature maps. We describe a new algorithm, called the CDF-Inversion(CDFI) Algorithm, that can be used to learn feature maps and, in the process, approximate an unknown probability distribution to within any specified accuracy. The primary advantages of the algorithm over previous feature-map algorithms are that it is simple enough to analyze mathematically for correctness and efficiency, and that it distributes the points of the map evenly, in a sense that can be made rigorous. Like other vector-quantization algorithms it is potentially useful for many applications, including monitoring and statistical modeling. While not a network algorithm, the CDFI algorithm is well-suited to implementation on parallel computers.
Article PDF
Similar content being viewed by others
References
Anderson, R., & Friedman, J. (1962).A limitation of the optimum property of the sequential probability ratio test.In Contributions to probability and statistics,6,Palo Alto,CA: Stanford University Press.
DeSieno, D. (1988).Adding a conscience to competitive learning.Proceedings of the Second International Conference on Neural Networks (pp.I-117-I-124).
Haussler, D. (1990).Generalizing the PAC model for neural net and other learning applications. (Technical Report UCSC-CRL-89-30).Santa Cruz,CA.: University of California.
Hecht-Nielsen, R. (1987).Counterpropagation networks.Proceedings of the First International Conference on Neural Networks.
Kohonen, T. (1982).Self-organized formation of topologically correct feature maps.Biological Cybernetics,43, 59-69.
Kohonen, T. (1984).Self-organization and associative memory.Springer-Verlag.
Kohonen, T. (1988).A "neural"phonetic typewriter.IEEE Computer,March,11-22.
Mood, J., & Graybill, F. (1963).Introduction to the theory of statistics (2nd edn.).New York: McGraw Hill.
Nasrabadi, N., & Feng, Y. (1988).Vector quantization of images based upon the Kohonen self-organizing feature maps.Proceedings of the First International Conference on Neural Networks (pp.I-101-I-106).
Paredis, J. (1989).Learning the behavior of dynamical systems from examples.Proceedings of the Sixth International Workshop on Machine Learning (pp.137-139).Morgan-Kaufmann.
Pollard, D. (1984).Convergence of stochastic processes.Springer-Verlag.
Ritter, H., & Schulten, K. (1986).On the stationary state of Kohonen's self-organizing sensory mapping.Biological Cybernetics,54,99-106.
Ritter, H., & Schulten, K. (1988).Kohonen's self-organizing maps:Exploring their computational capabilities. Proceedings of the Second International Conference on Neural Networks (pp.109-116).
Valiant, L.G. (1984).A theory of the learnable.C ACM,27,1134-1142.
Wald, A. (1947).Sequential analysis.New York: John Wiley.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Laird, P., Gamble, E. A “PAC” Algorithm for Making Feature Maps. Machine Learning 6, 145–160 (1991). https://doi.org/10.1023/A:1022654320034
Issue Date:
DOI: https://doi.org/10.1023/A:1022654320034