Unsupervised learning for neural trees
- 1 January 1991
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2709-2715 vol.3
- https://doi.org/10.1109/ijcnn.1991.170278
Abstract
A self-organizing neural tree is studied. The neural tree is suited to hierarchical classifications. Unsupervised learning algorithms have been developed for the neural tree. A simulation study indicated that the vectors represented by the nodes of the tree tend to approximate the probability of the sample distribution. The neural tree has been applied to speech recognition and image coding. Promising results have been obtained.Keywords
This publication has 15 references indexed in Scilit:
- Adaptive vector quantization using a self-development neural networkIEEE Journal on Selected Areas in Communications, 1990
- Connectionist learning proceduresArtificial Intelligence, 1989
- Nonlinear neural networks: Principles, mechanisms, and architecturesNeural Networks, 1988
- Vector quantization of images based upon the Kohonen self-organizing feature mapsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1988
- An introduction to computing with neural netsIEEE ASSP Magazine, 1987
- Competitive Learning: From Interactive Activation to Adaptive ResonanceCognitive Science, 1987
- Learning representations by back-propagating errorsNature, 1986
- How does a brain build a cognitive code?Psychological Review, 1980
- A General Theory of Classificatory Sorting Strategies: 1. Hierarchical SystemsThe Computer Journal, 1967
- A general theory of classificatory sorting strategies: II. Clustering systemsThe Computer Journal, 1967