The good news of the last chapter was that mixing or exchange behavior can be understood in quantitative terms. The bad news was that simple selectorecombinative GAs scale poorly on hard problems, largely the result of their inadequate mixing behavior. of course, this does not say that simple GAs are not useful; nor does it invalidate the success that many have had in using simple GAs to solve problems that are difficult to approach by other means. Moreover, the class of nearly decomposable problems with unknown decomposition is difficult to crack by any computational method. But the results presented in the last chapter do go a long way toward explaining the observed fiddling and twiddling with codings and operators that have long been a staple of practical GA application. It also makes us wonder whether it is possible to design what we have called competent GAs—GAs that solve boundedly difficulty problems quickly, reliably, and accurately without the need for problemspecific codings, operators, or other forms of human intervention.
- Building Block
- Linkage Group
- Probabilistic Expression
- Learn Classifier System
- Problem Difficulty
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
This is a preview of subscription content, access via your institution.
Tax calculation will be finalised at checkout
Purchases are for personal use onlyLearn about institutional subscriptions
Unable to display preview. Download preview PDF.
© 2002 Springer Science+Business Media Dordrecht
About this chapter
Cite this chapter
Goldberg, D.E. (2002). Design of Competent Genetic Algorithms. In: The Design of Innovation. Genetic Algorithms and Evolutionary Computation, vol 7. Springer, Boston, MA. https://doi.org/10.1007/978-1-4757-3643-4_12
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4757-3645-8
Online ISBN: 978-1-4757-3643-4
eBook Packages: Springer Book Archive