On improving genetic programming for symbolic regression

Citation

Gustafson S, Burke E & Krasnogor N (2005) On improving genetic programming for symbolic regression. In: 2005 IEEE Congress on Evolutionary Computation, IEEE CEC 2005. Proceedings. Vol. 1. The 2005 IEEE Congress on Evolutionary Computation, 2005, Edinburgh, Scotland, 05.09.2005-05.09.2005. Piscataway, NJ, USA: IEEE, pp. 912-919. http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=1554780; https://doi.org/10.1109/CEC.2005.1554780

Abstract
This paper reports an improvement to genetic programming (GP) search for the symbolic regression domain, based on an analysis of dissimilarity and mating. GP search is generally difficult to characterise for this domain, preventing well motivated algorithmic improvements. We first examine the ability of various solutions to contribute to the search process. Further analysis highlights the numerous solutions produced during search with no change to solution quality. A simple algorithmic enhancement is made that reduces these events and produces a statistically significant improvement in solution quality. We conclude by verifying the generalisability of these results on several other regression instances

Keywords
genetic algorithms; regression analysis; search problems