Abstract
Harik [47] took Holland’s call [53] for evolution of tight genetic linkage and proposed the linkage learning genetic algorithm (LLGA), which used a special probabilistic expression mechanism and a unique combination of the (gene number, allele) coding scheme and an exchange crossover operator to create an evolvable genotypic structure that made genetic linkage learning natural and viable for genetic algorithms. This integration of data structure and mechanism led to successful genetic linkage learning, particularly on problems composed of badly scaled building blocks. Interestingly, the nucleation procedure, which refers to the process of building-block formation, was less successful on problems with uniformly scaled building blocks, and this chapter seeks to better understand why this was so and to correct the deficiency by adopting a coding mechanism, promoters12, that exists in genetics.
Preview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Rights and permissions
About this chapter
Cite this chapter
Chen, Yp. A First Improvement: Using Promoters. In: Extending the Scalability of Linkage Learning Genetic Algorithms. Studies in Fuzziness and Soft Computing, vol 190. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11339380_6
Download citation
DOI: https://doi.org/10.1007/11339380_6
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-28459-8
Online ISBN: 978-3-540-32413-3
eBook Packages: EngineeringEngineering (R0)