Abstract
This chapter proposes a generic framework to build geometric dispersion (GD) operators for Geometric Semantic Genetic Programming in the context of symbolic regression, followed by two concrete instantiations of the framework: a multiplicative geometric dispersion operator and an additive geometric dispersion operator. These operators move individuals in the semantic space in order to balance the population around the target output in each dimension, with the objective of expanding the convex hull defined by the population to include the desired output vector. An experimental analysis was conducted in a testbed composed of sixteen datasets showing that dispersion operators can improve GSGP search and that the multiplicative version of the operator is overall better than the additive version.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The convex hull of a set of points is given by the set of all possible convex combinations of these points [18].
- 2.
Note that when \({\mathbf {x}}_i \in \mathbb {R}^d\) with d > 1 the vector I becomes a matrix with dimensions d × n. We allow an abuse of notation by representing the matrix as a vector with dimension n, where each element corresponds to a vector of dimension d.
- 3.
The coefficients of convex combinations can be found by means of Gaussian elimination [9].
- 4.
Oliveira et al. [15] presents the first geometric dispersion operator. However, this operator is a particular case of the framework presented in this paper. Hence, hereafter their operator is referred as multiplicative geometric dispersion (MGD) operator in contrast to the GD framework.
References
Albinati, J., Pappa, G.L., Otero, F.E.B., Oliveira, L.O.V.B.: The effect of distinct geometric semantic crossover operators in regression problems. In: Proceeding of the EuroGP, pp. 3–15 (2015)
Banzhaf, W., Nordin, P., Keller, R., Francone, F.: Genetic Programming—An Introduction: On the Automatic Evolution of Computer Programs and Its Applications. Morgan Kaufmann Publishers, San Francisco (1998)
Beadle, L., Johnson, C.G.: Semantic analysis of program initialisation in genetic programming. Genet. Program Evolvable Mach. 10(3), 307–337 (2009)
Botzheim, J., Cabrita, C., Kóczy, L.T., Ruano, A.E.: Genetic and bacterial programming for b-spline neural networks design. J. Adv. Comput. Intell. 11(2), 220–231 (2007)
Castelli, M., Vanneschi, L., Silva, S.: Semantic search-based genetic programming and the effect of intron deletion. IEEE Trans. Cybern. 44(1), 103–113 (2014)
Castelli, M., Silva, S., Vanneschi, L.: A C++ framework for geometric semantic genetic programming. Genet. Program Evolvable Mach. 16(1), 73–81 (2015)
Castelli, M., Trujillo, L., Vanneschi, L., Silva, S., Z-Flores, E., Legrand, P.: Geometric semantic genetic programming with local search. In: Proceeding of the GECCO’15, pp. 999–1006. ACM, New York (2015)
Castelli, M., Vanneschi, L., Silva, S., Ruberto, S.: How to exploit alignment in the error space: Two different GP models. In: Riolo, R., et al. (eds.) Genetic Programming Theory and Practice XII, Genetic and Evolutionary Computation, pp. 133–148. Springer International Publishing, Cham (2015)
Gentle, J.E.: Numerical Linear Algebra for Applications in Statistics. Statistics and Computing. Springer, New York (1998)
Gonçalves, I., Silva, S., Fonseca, C.M.F.: On the generalization ability of geometric semantic genetic programming. In: Proceeding of the EuroGP, pp. 41–52 (2015)
Koza, J.R.: Genetic Programming: On the Programming of Computers by Means of Natural Selection, vol. 1. MIT Press, Cambridge (1992)
Moraglio, A.: Abstract convex evolutionary search. In: Proceeding of the 11th FOGA, pp. 151–162 (2011)
Moraglio, A., Krawiec, K., Johnson, C.G.: Geometric semantic genetic programming. In: Proceeding of PPSN XII, vol. 7491, pp. 21–31. Springer, Berlin (2012)
Ni, J., Drieberg, R.H., Rockett, P.I.: The use of an analytic quotient operator in genetic programming. IEEE Trans. Evol. Comput. 17(1), 146–152 (2013)
Oliveira, L.O.V.B., Otero, F.E.B., Pappa, G.L.: A dispersion operator for geometric semantic genetic programming. In: Proceeding of the Genetic and Evolutionary Computation Conference, GECCO’16, pp. 773–780. ACM (2016)
Pawlak, T.P.: Competent algorithms for geometric semantic genetic programming. Ph.D. thesis, Poznan University of Technology, Poznan (2015)
Pawlak, T.P., Krawiec, K.: Semantic geometric initialization. In: Heywood, I.M., McDermott, J., Castelli, M., Costa, E., Sim, K. (eds.) Proceeding of the EuroGP’16. Lecture Notes in Computer Science, vol. 9594, pp. 261–277. Springer International Publishing, Cham (2016)
Roman, S.: Advanced Linear Algebra. Graduate Texts in Mathematics, vol. 135, 2nd edn. Springer, New York (2005)
Ruberto, S., Vanneschi, L., Castelli, M., Silva, S.: ESAGP - a semantic GP framework based on alignment in the error space. In: Proceeding of the EuroGP, pp. 150–161 (2014)
Acknowledgements
The authors would like to thank CAPES, FAPEMIG, and CNPq (141985/ 2015-1) for their financial support.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Oliveira, L.O.V.B., Otero, F.E.B., Pappa, G.L. (2018). A Generic Framework for Building Dispersion Operators in the Semantic Space. In: Riolo, R., Worzel, B., Goldman, B., Tozier, B. (eds) Genetic Programming Theory and Practice XIV. Genetic and Evolutionary Computation. Springer, Cham. https://doi.org/10.1007/978-3-319-97088-2_12
Download citation
DOI: https://doi.org/10.1007/978-3-319-97088-2_12
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-97087-5
Online ISBN: 978-3-319-97088-2
eBook Packages: Computer ScienceComputer Science (R0)