Centroid neural network for unsupervised competitive learning
- 1 March 2000
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks
- Vol. 11 (2) , 520-528
- https://doi.org/10.1109/72.839021
Abstract
An unsupervised competitive learning algorithm based on the classical -means clustering algorithm is proposed. The proposed learning algorithm called the centroid neural network (CNN) estimates centroids of the related cluster groups in training date. This paper also explains algorithmic relationships among the CNN and some of the conventional unsupervised competitive learning algorithms including Kohonen's self-organizing map (SOM) and Kosko's differential competitive learning (DCL) algorithm. The CNN algorithm requires neither a predetermined schedule for learning coefficient nor a total number of iterations for clustering. The simulation results on clustering problems and image compression problems show that CNN converges much faster than conventional algorithms with compatible clustering quality while other algorithms may give unstable results depending on the initial values of the learning coefficient and the total number of iterations.Keywords
This publication has 10 references indexed in Scilit:
- Self-organizing feature maps with self-adjusting learning parametersIEEE Transactions on Neural Networks, 1998
- Image compression by self-organized Kohonen mapIEEE Transactions on Neural Networks, 1998
- Topology preservation in self-organizing feature maps: exact definition and measurementIEEE Transactions on Neural Networks, 1997
- Image compression using self-organization networksIEEE Transactions on Circuits and Systems for Video Technology, 1994
- Differential competitive learning for centroid estimation and phoneme recognitionIEEE Transactions on Neural Networks, 1991
- The self-organizing mapProceedings of the IEEE, 1990
- Fast adaptive k-means clustering: some empirical resultsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1990
- Self-Organization and Associative MemoryPublished by Springer Nature ,1989
- Least squares quantization in PCMIEEE Transactions on Information Theory, 1982
- On learning and energy-entropy dependence in recurrent and nonrecurrent signed networksJournal of Statistical Physics, 1969