Abstract
Many GP learning methods have been proposed to decrease node combinations in order to keep the node combinations from explosively increasing. We propose a technique using an opposite approach which tests a greater number of combinations in order to decrease the chances of the search being ‘trapped’ in a local optimum. In the proposed technique, how ‘different’ the individual structure is is used as an index in selecting individuals for genetic operations. Therefore, variety in the GP group is strongly maintained, and it is expected that GP learning is always done to a new combination.
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
References
J. R. Koza, Genetic Programming, MIT Press, 1992
J. R. Koza, K. E. Kinner (ed.), et.al, Scalable Learning in Genetic Programming Using Automatic Function Definition, Advances in Genetic Programming, pp.99–117, 1994
A. Niimi, E. Tazaki, Extended Genetic Programming using pruning redundant patterns and fitting random continuous nodes, Proceedings of the 13th Annual Conference of JSAI, pp.257–258, 1999 (In Japanese)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1999 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Niimi, A., Tazaki, E. (1999). Rule Discovery Technique Using GP with Crossover to Maintain Variety. In: Arikawa, S., Furukawa, K. (eds) Discovery Science. DS 1999. Lecture Notes in Computer Science(), vol 1721. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-46846-3_42
Download citation
DOI: https://doi.org/10.1007/3-540-46846-3_42
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-66713-1
Online ISBN: 978-3-540-46846-2
eBook Packages: Springer Book Archive