Abstract
Coevolution of fitness predictors, which are a small sample of all training data for a particular task, was successfully used to reduce the computational cost of the design performed by cartesian genetic programming. However, it is necessary to specify the most advantageous number of fitness cases in predictors, which differs from task to task. This paper introduces a new type of directly encoded fitness predictors inspired by the principles of phenotypic plasticity. The size of the coevolved fitness predictor is adapted in response to the learning phase that the program evolution goes through. It is shown in 5 symbolic regression tasks that the proposed algorithm is able to adapt the number of fitness cases in predictors in response to the solved task and the program evolution flow.
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 subscriptionsReferences
Baldwin, J.M.: A new factor in evolution. Am. Nat. 30(354), 441–451 (1896)
Ellefsen, K.O.: Balancing the costs and benefits of learning ability. In: Advances in Artificial Life, ECAL 2013, vol. 12, pp. 292–299. MIT Press (2013)
Ellefsen, K.O.: Evolved sensitive periods in learning. In: Advances in Artificial Life, ECAL 2013, vol. 12, pp. 409–416. MIT Press (2013)
Hillis, W.D.: Co-evolving parasites improve simulated evolution as an optimization procedure. Physica D 42(1), 228–234 (1990)
Imamura, K., Foster, J.A., Krings, A.W.: The test vector problem and limitations to evolving digital circuits. In: Proceedings of the 2nd NASA/DoD Workshop on Evolvable Hardware, pp. 75–79. IEEE Computer Society (2000)
Jin, Y.: A comprehensive survey of fitness approximation in evolutionary computation. Soft Comput. J. 9(1), 3–12 (2005)
Miller, J.F.: Cartesian Genetic Programming. Springer, Berlin (2011)
Popovici, E., Bucci, A., Wiegand, R.P., de Jong, E.D.: Coevolutionary principles. In: Rozenberg, G., Bäck, T., Kok, J.N. (eds.) Handbook of Natural Computing, pp. 988–1028. Springer, New York (2011)
Schmidt, M.D., Lipson, H.: Coevolution of fitness predictors. IEEE Trans. Evol. Comput. 12(6), 736–749 (2008)
Sikulova, M., Hulva, J., Sekanina, L.: Indirectly encoded fitness predictors coevolved with cartesian programs. In: Machado, P., Heywood, M.I., McDermott, J., Castelli, M., García-Sánchez, P., Burelli, P., Risi, S., Sim, K. (eds.) Genetic Programming. LNCS, vol. 9025. Springer, Heidelberg (2015)
Sikulova, M., Sekanina, L.: Acceleration of evolutionary image filter design using coevolution in cartesian GP. In: Coello, C.A.C., Cutello, V., Deb, K., Forrest, S., Nicosia, G., Pavone, M. (eds.) PPSN 2012, Part I. LNCS, vol. 7491, pp. 163–172. Springer, Heidelberg (2012)
Šikulová, M., Sekanina, L.: Coevolution in cartesian genetic programming. In: Moraglio, A., Silva, S., Krawiec, K., Machado, P., Cotta, C. (eds.) EuroGP 2012. LNCS, vol. 7244, pp. 182–193. Springer, Heidelberg (2012)
Acknowledgements
This work was supported by the Czech Science Foundation project 14-04197S. The authors thank the IT4Innovations Centre of Excellence for enabling these experiments.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Wiglasz, M., Drahosova, M. (2016). Plastic Fitness Predictors Coevolved with Cartesian Programs. In: Heywood, M., McDermott, J., Castelli, M., Costa, E., Sim, K. (eds) Genetic Programming. EuroGP 2016. Lecture Notes in Computer Science(), vol 9594. Springer, Cham. https://doi.org/10.1007/978-3-319-30668-1_11
Download citation
DOI: https://doi.org/10.1007/978-3-319-30668-1_11
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-30667-4
Online ISBN: 978-3-319-30668-1
eBook Packages: Computer ScienceComputer Science (R0)