Abstract
In a recent contribution we have introduced a new implementation of geometric semantic operators for Genetic Programming. Thanks to this implementation, we are now able to deeply investigate their usefulness and study their properties on complex real-life applications. Our experiments confirm that these operators are more effective than traditional ones in optimizing training data, due to the fact that they induce a unimodal fitness landscape. Furthermore, they automatically limit overfitting, something we had already noticed in our recent contribution, and that is further discussed here. Finally, we investigate the influence of some parameters on the effectiveness of these operators, and we show that tuning their values and setting them “a priori” may be wasted effort. Instead, if we randomly modify the values of those parameters several times during the evolution, we obtain a performance that is comparable with the one obtained with the best setting, both on training and test data for all the studied problems.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Both in the text and in the pseudocode of Fig. 11.2, we abuse the term “ancestors” to designate not only the parents but also the random trees used to build an individual by crossover or mutation.
- 2.
Figure 11.1b makes an assumption: that the number of instances of the test set is identical to the one of the training set. Otherwise, the training target T and the test point τ could not be drawn in the same plane, because they would have different dimensions. This hypothesis is false in general, but it is used for simplicity, since it helps us to explain more clearly our hypothesis. Nevertheless, it is worth pointing out that, of course, the argument holds also when this restrictive assumption is false.
References
Archetti F, Lanzeni S, Messina E, Vanneschi L (2007) Genetic programming for computational pharmacokinetics in drug discovery and development. Genet Program Evolvable Mach 8:413–432
Beadle L, Johnson C (2008) Semantically driven crossover in genetic programming. In: Wang J (ed) Proceedings of the IEEE world congress on computational intelligence, Hong Kong. IEEE Computational Intelligence Society/IEEE, pp 111–116. doi:10.1109/CEC.2008.4630784
Beadle L, Johnson CG (2009) Semantically driven mutation in genetic programming. In: Tyrrell A (ed) 2009 IEEE congress on evolutionary computation, Trondheim. IEEE Computational Intelligence Society/IEEE, pp 1336–1342. doi:10.1109/CEC.2009.4983099
Koza JR (1992) Genetic programming: on the programming of computers by means of natural selection. MIT, Cambridge
Krawiec K (2012) Medial crossovers for genetic programming. In: Moraglio A, Silva S, Krawiec K, Machado P, Cotta C (eds) Proceedings of the 15th European conference on genetic programming, EuroGP 2012, Malaga. Lecture notes in computer science, vol 7244. Springer, pp 61–72. doi:10.1007/978-3-642-29139-5-6
McPhee NF, Ohs B, Hutchison T (2008) Semantic building blocks in genetic programming. In: Proceedings of the 11th European conference on genetic programming, EuroGP’08, Naples. Springer, Berlin/Heidelberg, pp 134–145
Moraglio A, Krawiec K, Johnson CG (2012) Geometric semantic genetic programming. In: Parallel problem solving from nature, PPSN XII (part 1), Taormina. Lecture notes in computer science, vol 7491. Springer, pp 21–31
Nguyen QU, Nguyen XH, O’Neill M (2009a) Semantic aware crossover for genetic programming: the case for real-valued function regression. In: Vanneschi L, Gustafson S, Moraglio A, De Falco I, Ebner M (eds) Proceedings of the 12th European conference on genetic programming, EuroGP 2009, Tuebingen. Lecture notes in computer science, vol 5481. Springer, pp 292–302. doi:10.1007/978-3-642-01181-8-25
Nguyen QU, Nguyen XH, O’Neill M (2009b) Semantics based mutation in genetic programming: the case for real-valued symbolic regression. In: Matousek R, Nolle L (eds) 15th international conference on soft computing, Mendel’09, Brno, pp 73–91
Uy NQ, Hoai NX, O’Neill M, McKay B (2010) The role of syntactic and semantic locality of crossover in genetic programming. In: Schaefer R, Cotta C, Kolodziej J, Rudolph G (eds) 11th international conference on parallel problem solving from nature, PPSN 2010, Krakow. Lecture notes in computer science, vol 6239. Springer, pp 533–542. doi:10.1007/978-3-642-15871-1-54
Uy NQ, Hoai NX, O’Neill M, McKay RI, Galvan-Lopez E (2011) Semantically-based crossover in genetic programming: application to real-valued symbolic regression. Genet Program Evolvable Mach 12(2):91–119. doi:10.1007/s10710-010-9121-2
Vanneschi L, Castelli M, Manzoni L, Silva S (2013) A new implementation of geometric semantic GP applied to predicting pharmacokinetic parameters. In: Proceedings of the 16th European conference on genetic programming, EuroGP’13, Vienna. Springer, pp 205–216
Acknowledgements
This work was supported by national funds through FCT under contract PEst-OE/EEI/LA0021/2013 and by projects EnviGP (PTDC/EIA-CCO/103363/2008), MassGP (PTDC/EEI-CTP/2975/2012) and InteleGen (PTDC/DTP-FTO/1747/2012), Portugal.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer Science+Business Media New York
About this chapter
Cite this chapter
Vanneschi, L., Silva, S., Castelli, M., Manzoni, L. (2014). Geometric Semantic Genetic Programming for Real Life Applications. In: Riolo, R., Moore, J., Kotanchek, M. (eds) Genetic Programming Theory and Practice XI. Genetic and Evolutionary Computation. Springer, New York, NY. https://doi.org/10.1007/978-1-4939-0375-7_11
Download citation
DOI: https://doi.org/10.1007/978-1-4939-0375-7_11
Published:
Publisher Name: Springer, New York, NY
Print ISBN: 978-1-4939-0374-0
Online ISBN: 978-1-4939-0375-7
eBook Packages: Computer ScienceComputer Science (R0)