Advertisement

Continuous Cartesian Genetic Programming with Particle Swarm Optimization

  • Jaroslav LoeblEmail author
  • Viera Rozinajová
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 941)

Abstract

Cartesian Genetic Programming (CGP) is a type of Genetic Programming, which uses a sequence of integers to represent an executable graph structure. The most common way of optimizing the CGP is to use a simple evolutionary strategy with mutations, which randomly changes the integer values of integer sequence. We propose an alternative genotype-phenotype mapping procedure for CGP allowing usage of real-valued numbers in genotype. Novel representation allows continuous transition between various functions and inputs of each given node (hence the name, Continuous CGP), which means, that the optimization of CGP individual is transformed from combinatorial optimization problem to continuous optimization problem. This allows leveraging various metaheuristic optimization algorithms. In this paper, we present results obtained by Particle Swarm Optimization algorithm, showing that continuous representation is able to outperform classic CGP in some benchmarks and provides competitive results with one of the best performing symbolic regression systems in literature.

Keywords

Cartesian Genetic Programming Particle Swarm Optimization Symbolic regression Evolutionary algorithms 

Notes

Acknowledgment

This work was partially supported by the Slovak Research and Development Agency under the contract APVV-16-0213 and by the Operational Programme Research & Innovation, funded by the ERDF, project No. ITMS 26240120039.

References

  1. 1.
    Al-Sahaf, H., Zhang, M., Al-Sahaf, A., Johnston, M.: Keypoints detection and feature extraction: a dynamic genetic programming approach for evolving rotation-invariant texture image descriptors. IEEE Trans. Evol. Comput. 21(6), 825–844 (2017)CrossRefGoogle Scholar
  2. 2.
    Al-Sahaf, H., Zhang, M., Johnston, M.: Binary image classification: a genetic programming approach to the problem of limited training instances. Evol. Comput. 24(1), 143–182 (2016)CrossRefGoogle Scholar
  3. 3.
    Biscani, F., Izzo, D.: esa/pagmo2: pagmo 2.9, August 2018.  https://doi.org/10.5281/zenodo.1406840
  4. 4.
    Castejón, F., Carmona, E.J.: Automatic design of analog electronic circuits using grammatical evolution. Appl. Soft Comput. 62, 1003–1018 (2018)CrossRefGoogle Scholar
  5. 5.
    Clegg, J.: Combining cartesian genetic programming with an estimation of distribution algorithm. In: Proceedings of the 10th Annual Conference on Genetic and Evolutionary Computation, GECCO 2008, pp. 1333–1334. ACM, New York (2008).  https://doi.org/10.1145/1389095.1389350
  6. 6.
    Clegg, J., Walker, J.A., Miller, J.F.: A new crossover technique for cartesian genetic programming. In: Proceedings of the 9th Annual Conference on Genetic and Evolutionary Computation, GECCO 2007, pp. 1580–1587. ACM, New York (2007).  https://doi.org/10.1145/1276958.1277276
  7. 7.
    Goldman, B.W., Punch, W.F.: Analysis of cartesian genetic programming’s evolutionary mechanisms. IEEE Trans. Evol. Comput. 19(3), 359–373 (2015).  https://doi.org/10.1109/TEVC.2014.2324539CrossRefGoogle Scholar
  8. 8.
    Holland, J.H.: Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence. MIT press, Cambridge (1992)CrossRefGoogle Scholar
  9. 9.
    Kalkreuth, R., Rudolph, G., Krone, J.: Improving convergence in cartesian genetic programming using adaptive crossover, mutation and selection. In: 2015 IEEE Symposium Series on Computational Intelligence, pp. 1415–1422, December 2015.  https://doi.org/10.1109/SSCI.2015.201
  10. 10.
    Kalkreuth, R., Rudolph, G., Droschinsky, A.: A new subgraph crossover for cartesian genetic programming. In: McDermott, J., Castelli, M., Sekanina, L., Haasdijk, E., García-Sánchez, P. (eds.) Genetic Programming, pp. 294–310. Springer International Publishing, Cham (2017)CrossRefGoogle Scholar
  11. 11.
    Kaufmann, P., Platzner, M.: Advanced techniques for the creation and propagation of modules in cartesian genetic programming. In: Proceedings of the 10th Annual Conference on Genetic and Evolutionary Computation, GECCO 2008, pp. 1219–1226. ACM, New York (2008).  https://doi.org/10.1145/1389095.1389334
  12. 12.
    Kennedy, J.: Particle Swarm Optimization, pp. 760–766. Springer, Boston (2010).  https://doi.org/10.1007/978-0-387-30164-8_630
  13. 13.
    Khan, M.M., Ahmad, A.M., Khan, G.M., Miller, J.F.: Fast learning neural networks using cartesian genetic programming. Neurocomputing 121, 274–289 (2013)CrossRefGoogle Scholar
  14. 14.
    Koza, J.R.: Genetic programming as a means for programming computers by natural selection. Stat. Comput. 4(2), 87–112 (1994)CrossRefGoogle Scholar
  15. 15.
    Lensen, A., Xue, B., Zhang, M.: GPGC: genetic programming for automatic clustering using a flexible non-hyper-spherical graph-based approach. In: Proceedings of the Genetic and Evolutionary Computation Conference, pp. 449–456. ACM (2017)Google Scholar
  16. 16.
    Meier, A., Gonter, M., Kruse, R.: Accelerating convergence in cartesian genetic programming by using a new genetic operator. In: Proceedings of the 15th Annual Conference on Genetic and Evolutionary Computation, GECCO 2013, pp. 981–988. ACM, New York (2013).  https://doi.org/10.1145/2463372.2463481
  17. 17.
    Miller, J.F., Thomson, P., Fogarty, T.: Designing electronic circuits using evolutionary algorithms. arithmetic circuits: a case study (1997)Google Scholar
  18. 18.
    Miller, J.F.: An empirical study of the efficiency of learning boolean functions using a cartesian genetic programming approach. In: Proceedings of the 1st Annual Conference on Genetic and Evolutionary Computation - Volume 2, GECCO 1999, pp. 1135–1142. Morgan Kaufmann Publishers Inc., San Francisco (1999)Google Scholar
  19. 19.
    Miller, J.F., Thomson, P.: Cartesian genetic programming. In: Poli, R., Banzhaf, W., Langdon, W.B., Miller, J., Nordin, P., Fogarty, T.C. (eds.) Genetic Programming, pp. 121–132. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  20. 20.
    Oesch, C.: P-tree programming. In: 2017 IEEE Symposium Series on Computational Intelligence (SSCI), pp. 1–7, November 2017. https://doi.org/10.1109/SSCI.2017.8280849
  21. 21.
    Papa, J.P., Rosa, G.H., Papa, L.P.: A binary-constrained geometric semantic genetic programming for feature selection purposes. Pattern Recognit. Lett. 100, 59–66 (2017)CrossRefGoogle Scholar
  22. 22.
    Poli, R., McPhee, N.F.: Covariant parsimony pressure in genetic programming. Technical report, Citeseer (2008)Google Scholar
  23. 23.
    Uy, N.Q., Hoai, N.X., O’Neill, M., McKay, R.I., Galván-López, E.: Semantically-based crossover in genetic programming: application to real-valued symbolic regression. Genet. Program. Evolvable Mach. 12(2), 91–119 (2011).  https://doi.org/10.1007/s10710-010-9121-2CrossRefGoogle Scholar
  24. 24.
    Voss, M.S.: Social programming using functional swarm optimization. In: Proceedings of the 2003 IEEE Swarm Intelligence Symposium, SIS 2003 (Cat. No.03EX706), pp. 103–109, April 2003.  https://doi.org/10.1109/SIS.2003.1202254
  25. 25.
    Walker, J.A., Miller, J.F., Cavill, R.: A multi-chromosome approach to standard and embedded cartesian genetic programming. In: Proceedings of the 8th Annual Conference on Genetic and Evolutionary Computation, GECCO 2006, pp. 903–910. ACM, New York (2006). https://doi.org/10.1145/1143997.1144153
  26. 26.
    White, D.R., McDermott, J., Castelli, M., Manzoni, L., Goldman, B.W., Kronberger, G., Jaśkowski, W., O’Reilly, U.M., Luke, S.: Better GP benchmarks: community survey results and proposals. Genet. Program. Evolvable Mach. 14(1), 3–29 (2013).  https://doi.org/10.1007/s10710-012-9177-2CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  1. 1.Faculty of Informatics and Information TechnologiesSlovak University of Technology in BratislavaBratislava 4Slovakia

Personalised recommendations