Journal of the Korean Institute of Intelligent Systems (한국지능시스템학회논문지)

Korean Institute of Intelligent Systems (한국지능시스템학회)
권호 표지
DOI QR코드

DOI QR Code

New Generation Gap Models for Evolutionary Algorithm in Real Parameter Optimization

실수최적화 진화 알고리즘을 위한 새로운 세대차 모델

  • Published : 2009.02.25
Download PDF
⟨ Previous Next ⟩

Abstract

Two new generation gap models with modified parent-centric recombination(PCX) operator are proposed. First, the self-adaptation generation gap(SGG) model is a control method that keeps a replaced probability of parents by offspring to a certain level which obtains better performance. Second, virtual cluster generation gap(VCGG) is provided to extend distances among parents using clustering, which causes it to diversify individuals. In this model, distances among parents can be controlled by size of clusters. To demonstrate the effectiveness of our two proposed approaches, experiments for three standard test problems are executed and compared to most competing current approaches, CMA-ES and Generalized Generation Gap(G3) with PCX. It is shown two proposed methods are superior to consistently other approaches in the study.

수정된 PCX(parent-centric recombination) 연산자와 결합한 두 가지 새로운 세대차 모델이 제안된다. 첫째, 자가적응 세대차 모델(SGG, self-adaptation generation gap)은 자손에 의한 부모의 대치 확률을 일정한 수준으로 유지하는 제어 방식이다. 둘째, 가상 클러스터 세대차(VCGG, virtual cluster generation gap)는 클러스터링을 통해 부모간의 거리를 조정해 주며, 이로 인해 개체들이 다양화 될 수 있다. 이 모델에서 부모간의 거리는 클러스터의 크기로 조절된다. 제안된 두 가지 접근법의 효용성을 입증하기 위해서 3 가지 표준적인 문제에 대한 실험이 수행되었다. 가장 최근의 경쟁력 있는 접근법인 CMA-ES와 G3-PCX와 비교한 결과, 제안된 두 기법 모두 기존의 접근법들 보다 우수함을 보여준다.

Keywords

References

  1. H.-P. Schwefel, Evolution and Optimum Seeking, John Wiley & Sons, Inc, 1995
  2. K. Deb, A. Anand and D. Joshi, A Computationally Efficient Evolutionary Algorithm Conference for Real-Parameter Optimization.' Evolutionary Computation. pp. 10(4): 371-395, 2002 https://doi.org/10.1162/106365602760972767
  3. I. Ono and S. Kobayashi, 'A real-coded genetic algorithm for function optimization using unimodal normal distribution crossover.' In Proceedings of the Seventh International Conference on Genetic Algorithms(ICGA-7), pp. 246-253, 1997
  4. T. Higuchi, S. Tsutsui, and M. Yamamura, 'Theoretical analysis of simplex crossover for real-coded genetic algorithms,' In Proceedings of Parallel Problem Solving from Nature (PPSN-VI), pp. 365-374, 2000
  5. K. Deb and R. B. Agrawal. 'Simulated binary crossover for continuous search space,' Complex Systems, 9(2) pp. 115-148, 1995
  6. H. Kita, I. Ono and S. Kobayashi, 'Multi-parental Extension of the Unimodal Normal Distribution Crossover for Real-Code Genetic Algorithm,' In Proceedings of IEEE Conference on Evolutionary Computation, Vol. 2 pp. 1581-1587, 1999
  7. H. Satoh, M. Yamamura, and S. Kobayashi, 'Minimal generation gap model for GAs considering both exploration and exploitation,' Methodologies for the conception, Design, and Application of Intelligent Systems, pp. 494-497, World Scientic, Singapore, 1996
  8. O. Takahashi, H. Kite, S. Kobayashi, 'A Distance Dependent Alternation on Real-coded Genetic Algorithms,' In Proceedings of IEEE Conference on Evolutionary Computation, pp. 619-624, Vol.1, 1999
  9. N. Hansen and A. Ostermeier, 'Adapting arbitrary normal mutation distributions in evolution strategies,' In Proceedings of IEEE Conference on Evolutionary Computation, pp. 312-317, 1996
  10. Sarma. J and De Jong. K, Generation gap methods. Hand book of Evolutionary Computation, pages C2.7:1-C2.7:5, Oxford University Press, UK, 1997
  11. M. M. Raghuwanshi and O. G. Kakde, 'Distributed Quasi Steady-State Genetic Algorithm with Niches and Species,' International Journal of Computational Intelligence Research, Vol. 3, No.2, pp. 155-164, 2007