Spatially Structured Evolutionary Algorithms: Graph Degree, Population Size and Convergence Speed

  • Carlos M. Fernandes
  • Juan L. J. Laredo
  • Agostinho C. Rosa
Chapter
Part of the Studies in Computational Intelligence book series (SCI, volume 737)

Abstract

An evolutionary algorithm (EA) is said to be spatially structured when its individuals are arranged in an incomplete graph and interact only with their neighbors. Previous studies argue that spatially structured EAs are less likely to converge prematurely to local optima. Furthermore, they have been initially designed for distributed computing and it is often claimed that their parallelization is simpler than the equivalent non-structured algorithm. However, most of the empirical studies on spatially structured EAs use a predefined and fixed population size, whereas the full potential of this or any other any kind of EA can only be explored if the population size is properly set. This paper investigates optimal population sizes of spatially structured EAs (cellular EAs, in particular) and the relationship between that size, convergence speed and the degree of the structuring network. EAs structured by regular graphs with different degrees have been tested on different types of fitness landscapes. We conclude that in most cases graphs with low degree require smaller populations to converge consistently to global optima. However, if the population size is properly set, EAs structured by graphs with higher degrees not only converge to global optima with high probability, but also converge faster.

Keywords

Evolutionary computation Spatially structured genetic algorithms Optimal population size Distributed EAs 

Notes

Acknowledgements

First author wishes to thank FCT, Ministério da Ciência e Tecnologia, his Research Fellowship SFRH/BPD/111065/2015). This work was supported by FCT PROJECT [UID/EEA/50009/2013].

References

  1. 1.
    Alba, E., Tomassini, M.: Parallelism and evolutionary algorithms. IEEE Trans. Evol. Comput. 6(5), 443–462 (2002)CrossRefGoogle Scholar
  2. 2.
    Alba, E., Dorronsoro, B.: The exploration/exploitation tradeoff in dynamic cellular genetic algorithms. IEEE Trans. Evol. Comput. 9, 126–142 (2005)CrossRefGoogle Scholar
  3. 3.
    Bäck, T.: Evolutionary Algorithms in Theory and Practice. Oxford University Press, Oxford (1996)MATHGoogle Scholar
  4. 4.
    Cantú-Paz, E.: Migration policies, selection pressure, and parallel EAs. Journal of Heuristics 7(4), 311–334 (2001)CrossRefMATHGoogle Scholar
  5. 5.
    Fernandes, C.M., Laredo, J.L.J., Merelo, J.J., Cotta, C., Rosa, A.C.: Dynamic and Partially Connected Ring Topologies for Evolutionary Algorithms with Structured Populations, EvoApplications 2014: Applications of Evolutionary Computation, pp. 665–677 (2014)Google Scholar
  6. 6.
    Giacobini, M., Tomassini, M., Tettamanzi, A.: Takeover time curves in random and small-world structured populations. In: Proceedings of the 7th GECCO, pp. 1333–1340 (2005)Google Scholar
  7. 7.
    Giacobini, M., Tomassini, M., Tettamanzi, A.G.B., Alba, E.: Selection intensity in cellular evolutionary algorithms for regular lattices. IEEE Trans. Evol. Comput. 9, 489–505 (2005)CrossRefGoogle Scholar
  8. 8.
    Laredo, J.L.J., Bouvry, P., González, D.L., Fernandéz de la Vega, F., Arenas, M.G., Merelo, J.J., Fernandes, C.M.: Designing robust volunteer-based evolutionary algorithms. Genet. Program Evol. Mach. 15(3), 221–244 (2014)CrossRefGoogle Scholar
  9. 9.
    Payne, J.L, Eppstein, M.J.: Emergent mating topologies in spatially structured genetic algorithms. In: Proceedings of 8th GECCO, pp. 207–214 (2006)Google Scholar
  10. 10.
    Réka, A., Barabási, A.-L.: Statistical mechanics of complex networks. Rev. Mod. Phys. 74, 47–94 (2000)MathSciNetMATHGoogle Scholar
  11. 11.
    Sarma J., De Jong, K.: An analysis of the effect of the neighborhood size and shape on local selection algorithms. In: Proceedings of International Conference on Parallel Problem Solving from Nature IV, LNCS 1141, pp. 236–244. Springer (1996)Google Scholar
  12. 12.
    Sastry, K.: Evaluation-relaxation schemes for genetic and evolutionary algorithms. M.Sc. thesis, University of Illinois, Urbana, IL, USA (2001)Google Scholar
  13. 13.
    Tomassini, M.: Spatially Structured Evolutionary Algorithms. Springer, Heidelberg (2005)MATHGoogle Scholar
  14. 14.
    Whitacre, J.M., Sarker, R.A., Pham, Q.: The self-organization of interaction networks for nature-inspired optimization. IEEE Trans. Evol. Comput. 12, 220–230 (2008)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  • Carlos M. Fernandes
    • 1
    • 2
  • Juan L. J. Laredo
    • 3
  • Agostinho C. Rosa
    • 1
  1. 1.LARSyS: Laboratory for Robotics and Systems in Engineering and ScienceUniversity of LisbonLisbonPortugal
  2. 2.Department of Computer ArchitectureUniversity of GranadaGranadaSpain
  3. 3.LITIS, University of Le HavreLe HavreFrance

Personalised recommendations