A Framework for Agent-Based Evaluation of Genetic Algorithms
Genetic Algorithms (GA) are a set of algorithms that use biological evolution as inspiration to solve search problems. One of the difficulties found when working with GA are the several parameters that have to be set and the many details that can be tunned in the GA. Usually it leads to the execution of several experiments in order to study how the GA behaves under different circumstances. In general it requires several computational resources and time to code the same algorithm with slight differences several times. In this paper we propose a framework based on agent technology able to parallelize the experiment and to split it into several components. It is complemented with a description of how this framework can be used in the evolution of regular expressions.
KeywordsGenetic Algorithm Crossover Operator Regular Expression Chromosome Length Grammatical Evolution
Unable to display preview. Download preview PDF.
- 1.Barrero, D.F., Camacho, D., R-Moreno, M.D.: Automatic Web Data Extraction Based on Genetic Algorithms and Regular Expressions. In: Data Mining and Multiagent Integration, August 2009. Springer, Heidelberg (2009)Google Scholar
- 2.Barrero, D.F., R-Moreno, M.D., López, D.R., García, Ó.: Searchy: A metasearch engine for heterogeneus sources in distributed environments. In: Proceedings of the International Conference on Dublin core and Metadata Applications, Madrid, Spain, September 2005, pp. 261–265 (2005)Google Scholar
- 3.Chu, D., Rowe, J.E.: Crossover operators to control size growth in linear GP and variable length GAs. In: Wang, J. (ed.) 2008 IEEE World Congress on Computational Intelligence, Hong Kong, June 1-6. IEEE Computational Intelligence Society. IEEE Press, Los Alamitos (2008)Google Scholar
- 4.Deb, K.: Binary and floating-point function optimization using messy genetic algorithms. PhD thesis, Tuscaloosa, AL, USA (1991)Google Scholar
- 6.Harvey, I.: The saga cross: the mechanics of recombination for species with variablelength genotypes. In: Manner, R., Manderick, B. (eds.) Parallel Problem, pp. 269–278. North-Holland, Amsterdam (1992)Google Scholar
- 7.Koza, J.R.: Genetic Programming: On the Programming of Computers by Means of Natural Selection (Complex Adaptive Systems). The MIT Press, Cambridge (1992)Google Scholar
- 9.Parekh, R., Honavar, V.: Grammar inference, automata induction, and language acquisition. In: Handbook of Natural Language Processing, pp. 727–764. Marcel Dekker, New York (1998)Google Scholar
- 10.Rana, S.: The distributional biases of crossover operators. In: Proceedings of the Genetic and Evolutionary Computation Conference, pp. 549–556. Morgan Kaufmann Publishers, San Francisco (1999)Google Scholar
- 12.Spears, W.M.: Crossover or mutation. In: Foundations of Genetic Algorithms 2, pp. 221–237. Morgan Kaufmann, San Francisco (1993)Google Scholar