Abstract
Recently, it has been stated that the complexity of a solution is a good indicator of the amount of overfitting it incurs. However, measuring the complexity of a program, in Genetic Programming, is not a trivial task. In this paper, we study the functional complexity and how it relates with overfitting on symbolic regression problems. We consider two measures of complexity, Slope-based Functional Complexity, inspired by the concept of curvature, and Regularity-based Functional Complexity based on the concept of Hölderian regularity. In general, both complexity measures appear to be poor indicators of program overfitting. However, results suggest that Regularity-based Functional Complexity could provide a good indication of overfitting in extreme cases.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Jaffard, S.: Wavelet techniques in multifractal analysis. In: Fractal Geometry and Applications: A Jubilee of Benoit Mandelbrot, Proceedings of Symposium in Pure Mathematics, vol. 72, pp. 91–151 (2004)
Jones, T., Forrest, S.: Fitness distance correlation as a measure of problem difficulty for genetic algorithms. In: Proceedings of the Sixth International Conference on Genetic Algorithms, pp. 184–192. Morgan Kaufmann, San Francisco (1995)
Legrand, P., Lévy-Véhel, J.: Local regularity-based interpolation. In: WAVELET X, Part of SPIE’s Symposium on Optical Science and Technology, vol. 5207 (2003)
Lévy-Véhel, J., Legrand, P.: Signal and Image Processing with FRACLAB. In: Thinking in Patterns, pp. 321–322 (2004), http://fraclab.saclay.inria.fr/homepage.html
McDermott, J., Galvan-Lopez, E., O’Neill, M.: A fine-grained view of GP locality with binary decision diagrams as ant phenotypes. In: Schaefer, R., Cotta, C., Kołodziej, J., Rudolph, G. (eds.) PPSN XI. LNCS, vol. 6238, pp. 164–173. Springer, Heidelberg (2010)
Mitchell, T.: Machine Learning. McGraw-Hill, New York (1997)
Morvan, J.-M.: Generalized Curvatures, 1st edn. Springer, Heidelberg (2008)
Rosca, J.: Generality versus size in genetic programming. In: Proceedings of Genetic Programming 1996, pp. 381–387. MIT Press, Cambridge (1996)
Silva, S., Almeida, J.: Gplab: A genetic programming toolbox for matlab. In: Proceedings of the Nordic MATLAB Conference, pp. 273–278 (2003)
Silva, S., Costa, E.: Dynamic limits for bloat control in genetic programming and a review of past and current bloat theories. Genetic Programming and Evolvable Machines 10(2), 141–179 (2009)
Tricot, C.: Curves and Fractal Dimension. Springer, Heidelberg (1995)
Trujillo, L., Legrand, P., Lévy-Véhel, J.: The estimation of hölderian regularity using genetic programming. In: Proceedings of GECCO 2010, pp. 861–868. ACM, New York (2010)
Trujillo, L., Legrand, P., Olague, G., Pérez, C.: Optimization of the hölder image descriptor using a genetic algorithm. In: Proceedings of GECCO 2010, pp. 1147–1154. ACM, New York (2010)
Vanneschi, L., Castelli, M., Silva, S.: Measuring bloat, overfitting and functional complexity in genetic programming. In: Proceedings of GECCO 2010, pp. 877–884. ACM, New York (2010)
Vladislavleva, E.J., Smits, G.F., Den Hertog, D.: Order of nonlinearity as a complexity measure for models generated by symbolic regression via pareto genetic programming. IEEE Trans. Evol. Comp. 13, 333–349 (2009)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Trujillo, L., Silva, S., Legrand, P., Vanneschi, L. (2011). An Empirical Study of Functional Complexity as an Indicator of Overfitting in Genetic Programming. In: Silva, S., Foster, J.A., Nicolau, M., Machado, P., Giacobini, M. (eds) Genetic Programming. EuroGP 2011. Lecture Notes in Computer Science, vol 6621. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-20407-4_23
Download citation
DOI: https://doi.org/10.1007/978-3-642-20407-4_23
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-20406-7
Online ISBN: 978-3-642-20407-4
eBook Packages: Computer ScienceComputer Science (R0)