ABSTRACT
Models of Genetic Programming (GP) frequently reflect a neo-Darwinian view to evolution in which inheritance is based on a process of gradual refinement and the resulting solutions take the form of single monolithic programs. Conversely, introducing an explicitly symbiotic model of inheritance makes a divide-and-conquer metaphor for problem decomposition central to evolution. Benchmarking gradualist versus symbiotic models of evolution under a common evolutionary framework illustrates that not only does symbiosis result in more accurate solutions, but the solutions are also much simpler in terms of instruction and attribute count over a wide range of classification problem domains.
- S. Ali and K.A. Smith. On learning algorithm selection for classification. Applied Soft Computing, 6:119--138, 2006. Google ScholarDigital Library
- E.B. Baum and I.Durdanovic. Toward code evolution by artificial economies. In Evolution as Computation, pages 314--332. Springer, 2002.Google ScholarCross Ref
- M. Brameier and W. Banzhaf. A comparison of linear genetic programming and neural networks if medical data mining. IEEE Transactions on Evolutionary Computation, 5(1):17--26. 2001. Google ScholarDigital Library
- S. Brave. Evolving recursive programs for tree search. In P.J. Angeline and K.E. Kinnear, editors, Advances in Genetic Programming, volume 2, chapter 10, pages 203--220. MIT, 1996. Google ScholarDigital Library
- J. M. Daida, C.S. Grasso, S. A. Stanhope, and S. J. Ross. Symbionticism and complex adaptive systems I: Implications of having symbiosis occur in nature. In Proceedings of the Annual Conference on Evolutionary Programming, pages 177--186. MIT Press, 1996.Google Scholar
- E. D. de Jong. A monolithic archive for Pareto-coevolution. Evolutionary Computation, 15(1):61--94, 2007. Google ScholarDigital Library
- J. Doucette, P.Lichodzijewski, and M.I. Heywood. Evolving coevolutionary classifiers under large attribute spaces. In Genetic Programming Theory and Practice VII, pages 37--54. Springer, 2009.Google Scholar
- I. Guyon, S. Gunn, M. Nikravesh, and L. Zadeh, editors. Feature Extraction: Foundations and Applications, volume 207 of Studies in Fuzziness and Soft Computing. Springer, 2006. Google ScholarDigital Library
- M.I. Heywood and P.Lichodzijewski. Symbiogenesis as a mechanism for building complex adaptive systems: A review. In EvoApplications, volume 6024 of LNCS, pages 51--60. 2010. Google ScholarDigital Library
- L. Huelsbergen. Learning recursive sequences via evolution of machine-language programs. In Proceedings of the Annual Conference on Genetic Programming. pages 186--194, 1997.Google Scholar
- J. R. Koza. Genetic Programming: On the programming of computers by means of natural selection MIT, 1992. Google ScholarDigital Library
- U. Kutschera. Symbiogenesism natural selection, and the dynamic Earth. Theory in Biosciences, 128:191--203, 2009.Google ScholarCross Ref
- M. Lemczyk and M. I. Heywood. Training binary GP classifiers efficiently: A Pareto-coevolutionary approach. In European Conference on Genetic Programming, volume 4445 of LNCS, pages 229--240, 2007. Google ScholarDigital Library
- P. Lichodzijewski and M. I. Heywood. GP classifier problem decomposition using first-price and second-price auctions. In European Conference on Genetic Programming, volume 4445 of LNCS, pages 137--147, 2007. Google ScholarDigital Library
- P. Lichodzijewski and M. I. Heywood. Managing team-based problem solving with symbiotic bid-based genetic programming. In Proceedings of the Genetic and Evolutionary Computation Conference, pages 363--370, 2008. Google ScholarDigital Library
- Y. Liu, X.Yao, and T.Higuchi. Evolutionary Ensembles with negative correlation learning. IEEE Transactions on Evolutionary Computation, 4(4): 380--387. 2000. Google ScholarDigital Library
- L. Margulis and R.Fester, editors. Symbiosis as a Source of Evolutionary Innovation. MIT Press, 1991.Google Scholar
- J. Maynard Smith and E. Szathmary. The origins of life. Oxford University Press, 1999.Google Scholar
- A. R. McIntyre and M. I. Heywood. Pareto cooperative-competitive Genetic Programming: A classification benchmarking study. In Genetic Programming Theory and Practice VI, pages 43--60. Springer, 2008.Google Scholar
- D. E. Moriarty and R. Miikkulainen. Forming neural networks through efficient and adaptive coevolution. Evolutionary Computation, 5(4):373--399, 1998. Google ScholarDigital Library
- L. Panait, S. Luke, and R. P. Wiegand. Biasing coevolutionary search for optimal multiagent behaviors. IEEE Transactions on Evolutionary Computation, 10(6):629--645. 2006. Google ScholarDigital Library
- M. Potter and K. De Jong. Cooperative coevolution: an architecture for evolving coadapted subcomponents. Evolutionary Computation, 8(1):1--29, 2000. Google ScholarDigital Library
- J. P. Rosca and D. H. Ballard. Hierarchical self-organization in genetic programming. In International Conference on Machine Learning, page. 251--258, 1994.Google ScholarCross Ref
- R. Thomason and T. Soule. Novel ways of improving cooperation and performance in ensemble classifiers. In Proceedings of the Genetic and Evolutionary Computation Conference, pages 1708--1715, 2007. Google ScholarDigital Library
- S. van Dijk, D.Thierens, and M. de Berg. On the design and analysis of competent selecto-recominative GAs. Evolutionary Computation, 12(2): 243--267, 2004.. Google ScholarDigital Library
Index Terms
- Symbiosis, complexification and simplicity under GP
Recommendations
HMXT-GP: an information-theoretic approach to genetic programming that maintains diversity
SAC '11: Proceedings of the 2011 ACM Symposium on Applied ComputingThis paper applies a recent information--theoretic approach to controlling Genetic Algorithms (GAs) called HMXT to tree--based Genetic Programming (GP). HMXT, in a GA domain, requires the setting of selection thresholds in a population and the ...
Symbiosis Promotes Fitness Improvements in the Game of Life
We present a computational simulation of evolving entities that includes symbiosis with shifting levels of selection. Evolution by natural selection shifts from the level of the original entities to the level of the new symbiotic entity. In the simulation,...
Symbiotic coevolutionary genetic programming: a benchmarking study under large attribute spaces
Classification under large attribute spaces represents a dual learning problem in which attribute subspaces need to be identified at the same time as the classifier design is established. Embedded as opposed to filter or wrapper methodologies address ...
Comments