Adaptation, Performance and Vapnik-Chervonenkis Dimension of Straight Line Programs
We discuss here empirical comparation between model selection methods based on Linear Genetic Programming. Two statistical methods are compared: model selection based on Empirical Risk Minimization (ERM) and model selection based on Structural Risk Minimization (SRM). For this purpose we have identified the main components which determine the capacity of some linear structures as classifiers showing an upper bound for the Vapnik-Chervonenkis (VC) dimension of classes of programs representing linear code defined by arithmetic computations and sign tests. This upper bound is used to define a fitness based on VC regularization that performs significantly better than the fitness based on empirical risk.
KeywordsGenetic Programming Linear Genetic Programming Vapnik-Chervonenkis dimension
Unable to display preview. Download preview PDF.
- 2.Alonso, C.L., Montaña, J.L., Puente, J.: Straight line programs: a new Linear Genetic Programming Approach. In: Proc. 20th IEEE International Conference on Tools with Artificial Intelligence, ICTAI, pp. 517–524 (2008)Google Scholar
- 5.Teytaud, O., Gelly, S., Bredeche, N., Schoenauer, M.A.: Statistical Learning Theory Approach of Bloat. In: Proceedings of the 2005 conference on Genetic and Evolutionary Computation, pp. 1784–1785 (2005)Google Scholar
- 9.Lugosi, G.: Pattern classification and learning theory. In: Principles of Nonparametric Learning, pp. 5–62. Springer, Heidelberg (2002)Google Scholar