Abstract
Co-evolution is the term used to identify the process in nature in which two or more species interact so intimately that their evolutionary fitness depends on each other. Biological co-evolution has been the inspiration for a class of computational algorithms called co-evolutionary computing. Co-evolutionary design is an approach to design problem solving in which the requirements and solutions of design evolve separately and affect each other. A reconsideration of the purpose of the fitness function and its affect on convergence is necessary since the fitness function changes through the co-evolutionary cycles. The interactions between requirements and solutions of design may possibly add some new variables to both aspects of design, which redefines the search space for requirements and solutions as well as the fitness function. Based on the idea of mutualism, which is one of three types of co-evolution in nature, the interacting populations raise the level of fitness in both, rather than the two populations competing with each other or one population living off the other.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
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
Maher, M.L., Wu, P.X. (1999). Reconsidering Fitness and Convergence in Co-evolutionary Design. In: Foo, N. (eds) Advanced Topics in Artificial Intelligence. AI 1999. Lecture Notes in Computer Science(), vol 1747. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-46695-9_50
Download citation
DOI: https://doi.org/10.1007/3-540-46695-9_50
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-66822-0
Online ISBN: 978-3-540-46695-6
eBook Packages: Springer Book Archive