Improving the Parsimony of Regression Models for an Enhanced Genetic Programming Process
This research is focused on reducing the average size of the solutions generated by an enhanced GP process without affecting the high predictive accuracy the method exhibits when being applied on a complex, industry proposed, regression problem. As such, the effects the GP enhancements have on bloat have been studied and, finally, a bloat control system based on dynamic depth limiting (DDL) and iterated tournament pruning (ITP) was designed. The resulting bloat control system is able to improve by ≃ 40% the average GP solution parsimony without impacting average solution accuracy.
Keywordsgenetic programming symbolic regression solution parsimony bloat control
Unable to display preview. Download preview PDF.
- 2.Crawford-Marks, R., Spector, L.: Size control via size fair genetic operators in the PushGP genetic programming system. In: GECCO 2002: Proceedings of the Genetic and Evolutionary Computation Conference, July 9-13, pp. 733–739. Morgan Kaufmann Publishers, New York (2002)Google Scholar
- 6.Kronberger, G., Feilmayr, C., Kommenda, M., Winkler, S., Affenzeller, M., Burgler, T.: System identification of blast furnace processes with genetic programming. In: Logistics and Industrial Informatics - LINDI, pp. 1–6. IEEE Press, Los Alamitos (2009)Google Scholar
- 9.Winkler, S.M.: Evolutionary System Identification. Ph.D. thesis, Johannes-Kepler-Universität, Linz, Austria (2008)Google Scholar
- 11.Zăvoianu, A.C.: Towards solution parsimony in an enhanced genetic programming process. Master’s thesis, International School Informatics: Engineering & Management, ISI-Hagenberg, Johannes Kepler University, Linz (2010)Google Scholar