Abstract
The stochastic competitive evolutionary neural tree (SCENT) is a new unsupervised neural net that dynamically evolves a representational structure in response to its training data. Uniquely SCENT requires no initial parameter setting as it autonomously creates appropriate parameterisation at runtime. Pruning and convergence are stochastically controlled using locally calculated heuristics. A thorough investigation into the performance of SCENT is presented. The network is compared to other dynamic tree based models and to a high quality flat clusterer over a variety of data sets and runs.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
K. Butchart, N. Davey, and R. Adams. Hierarchical classification with a competitive evolutionary neural tree. Neural Networks, 1995. to appear.
K. Butchart, N. Davey, and R. Adams. A Comparative Study of two Self Organising and Structually Adaptive Dynamic NeuralTree Networks. John Wiley, Chichester, 1996.
K. Butchart, N. Davey, and R. Adams. Hierarchical classification with a stochastic competitive evolutionary neural tree. In Proceedings of ICNN96, volume 2, pages 1372–1377, 1996.
B.S. Everit. Cluster Analysis. Edward Arnold, London, 1993.
J. Hertz, A. Krogh, and R. Palmer. Introduction to the theory of Neural Computation. Addison Wesley, New Jersey, 1991.
T. Li, Y. Tan, S. Suen, and L. Fang. A structurally adaptive neural tree for recognition of a large character set. In Proc. 11th IAPR International Joint Conference on Pattern Recognition, volume II, pages 187–190, 1992.
T. Martinetz, S. Berkovich, and K. Schulten. Neural-gas network for vector quantisation and its application to time-series prediction. IEEE transactions on Neural Networks, 44(4), 1993.
J. Racz and T. Klotz. Knowledge representation by dynamic competitive learning techniques. SPIE Applications of Artificial Neural Networks II, 1469:778–783, 1991.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1998 Springer-Verlag Wien
About this paper
Cite this paper
Butchart, K., Davey, N., Adams, R.G. (1998). An Investigation into the Performance and Representations of a Stochastic, Evolutionary Neural Tree. In: Artificial Neural Nets and Genetic Algorithms. Springer, Vienna. https://doi.org/10.1007/978-3-7091-6492-1_122
Download citation
DOI: https://doi.org/10.1007/978-3-7091-6492-1_122
Publisher Name: Springer, Vienna
Print ISBN: 978-3-211-83087-1
Online ISBN: 978-3-7091-6492-1
eBook Packages: Springer Book Archive