Demonstrating the Evolution of Complex Genetic Representations: An Evolution of Artificial Plants
- 262 Downloads
A common idea is that complex evolutionary adaptation is enabled by complex genetic representations of phenotypic traits. This paper demonstrates how, according to a recently developed theory, genetic representations can self-adapt in favor of evolvability, i.e., the chance of adaptive mutations. The key for the adaptability of genetic representations is neutrality inherent in non-trivial genotype-phenotype mappings and neutral mutations that allow for transitions between genetic representations of the same phenotype. We model an evolution of artificial plants, encoded by grammar-like genotypes, to demonstrate this theory.
KeywordsType Mutation Phenotypic Variability Direct Encode Search Distribution Organism State
Unable to display preview. Download preview PDF.
- 1.L. Altenberg. Genome growth and the evolution of the genotype-phenotype map. In W. Banzhaf and F. H. Eeckman, editors, Evolution and Biocomputation: Computational Models of Evolution, pages 205–259. Springer, Berlin, 1995.Google Scholar
- 2.T. Bäck. Evolutionary Algorithms in Theory and Practice. Oxford University Press, 1996.Google Scholar
- 4.T. F. Hansen and G. P. Wagner. Epistasis and the mutation load: A measurementtheoretical approach. Genetics, 158:477–485, 2001.Google Scholar
- 6.G. S. Hornby and J. B. Pollack. The advantages of generative grammatical encodings for physical design. In Proceedings of the 2001 Congress on Evolutionary Computation (CEC 2001), pages 600–607. IEEE Press, 2001.Google Scholar
- 9.S. Lucas. Growing adaptive neural networks with graph grammars. In Proc. of European Symp. on Artificial Neural Netw. (ESANN 1995), pages 235–240, 1995.Google Scholar
- 10.M. Pelikan, D. E. Goldberg, and F. Lobo. A survey of optimization by building and using probabilistic models. Technical Report IlliGAL-99018, Illinois Genetic Algorithms Laboratory, 1999.Google Scholar
- 14.M. Toussaint. On the evolution of phenotypic exploration distributions. In C. Cotta, K. De Jong, R. Poli, and J. Rowe, editors, Foundations of Genetic Algorithms 7 (FOGA VII). Morgan Kaufmann, 2003. In press.Google Scholar
- 18.R. Watson and J. Pollack. A computational model of symbiotic composition in evolutionary transitions. Biosystems, Special Issue on Evolvability, 2002.Google Scholar