DOI QR코드

DOI QR Code

Improvement of the GA's Convergence Speed Using the Sub-Population

보조 모집단을 이용한 유전자 알고리즘의 수렴속도 개선

  • Lee, Hong-Kyu (School of Electrical, Electronics and Communication Engineering, Korea University of Technology and Education) ;
  • Lee, Jae-Oh (School of Electrical, Electronics and Communication Engineering, Korea University of Technology and Education)
  • 이홍규 (한국기술교육대학교 전기전자통신공학부) ;
  • 이재오 (한국기술교육대학교 전기전자통신공학부)
  • Received : 2014.07.03
  • Accepted : 2014.10.10
  • Published : 2014.10.31

Abstract

Genetic Algorithms (GAs) are efficient methods for search and optimization problems. On the other hand, there are some problems associated with the premature convergence to local optima of the multimodal function, which has multi peaks. The problem is related to the lack of genetic diversity of the population to cover the search spaces sufficiently. A sharing and crowding method were introduced. This paper proposed strategies to improve the convergence speed and the convergence to the global optimum for solving the multimodal optimization function. These strategies included the random generated sub-population that were well-distributed and spread widely through search spaces. The results of the simulation verified the effects of the proposed method.

유전자 알고리즘은 탐색과 최적화 문제에 대한 효과적인 방법으로 이용되고 있으나 다수의 정점이 있는 다중정점 함수에 대한 응용에 있어서는 지역해에 조기 수렴하여 고착되는 등 전역 최적해를 찾는데 어려움이 있다. 이러한 문제는 탐색공간을 충분히 탐색할 수 있는 모집단의 다양성이 부족한 데 기인하는 것이며 해결방법으로 니칭 방법과 크라우딩 방법 등이 소개되고 있다. 개체군의 다양성을 증가시키는 방법으로 지역해에 고착되지 않고 전역 최적해로 수렴되도록 하는 데 기본을 두고 있다. 본 논문에서는 다중정점 함수의 전역 최적해에 수렴하고 수렴속도를 높이는 방법으로 진화과정의 매 세대마다 탐색영역에 충분히 분포되도록 임의로 생성된 보조 모집단을 공급함으로서 안정적으로 전역 최적해로 수렴하는 방법을 제안하였다. 컴퓨터 모의실험을 통하여 본 논문에서 제안한 방법을 입증하였다.

Keywords

References

  1. Hong K. Lee, D.-H. Lee, Z. Ran, G. Lee, and M. Lee, "On Parameter Selection for Reducing Premature Convergence of Genetic Algorithms," in Proceeding of 23rd International Conference on Computer and Their Applications in Industry and Engineering, Las Vegas: NV, pp. 214-219, Nov. 2010.
  2. Hong K. Lee, "On Sweeping Operator for Reducing Premature Convergence of Genetic Algorithms," Journal of Institute of Control, Robotics and Systems, Vol. 17, No. 12, pp. 1210-1218, Dec. 2011. DOI: http://dx.doi.org/10.5302/J.ICROS.2011.17.12.1210
  3. Hong K. Lee, "Hybrid Genetic Operators of Hamming Distance and Fitness for Reducing Premature Convergence," The Journal of Korea Navigation Institute, Vol. 18, No. 2, pp. 170-177, Apr. 2014. DOI: http://dx.doi.org/10.12673/jant.2014.18.2.170
  4. D. Goldberg and J. Richardson, "Genetic algorithms with sharing for multimodal function optimization," in Proceeding of the 2nd International Conference on Genetic Algorithms and their Applications, Hilsdale, NJ, pp. 41-49, July, 1987.
  5. S.-H. Bae and B.-R. Moon, "Mutation rates in the context of hybrid genetic algorithms", in Proceeding of the Genetic and Evolutionary Computation Conference, Seattle: WA, pp. 381-382, June, 2004.
  6. M. Srinivas and M. Patnaik, "Adaptive probabilities of crossover and mutation in genetic algorithms," IEEE Trans. on Systems, Man and Cybernetics, Vol. 24, No. 4. pp. 656-667, Apr. 1994. DOI: http://dx.doi.org/10.1109/21.286385
  7. W. Cedeno, V. Vemuri, and T. Slezak, "Multi-niche crowding in genetic algorithms and its application to the assembly of DNA restriction-fragments," Journal of Evolutionary Computation, Vol. 2, No. 4, pp. 321-345, 1995. DOI: http://dx.doi.org/10.1162/evco.1994.2.4.321
  8. B. Sareni and L. Krahenbuhl, "Fitness Sharing and Niching Methods Revisited," IEEE Trans. on Evolutionary Computation, Vol. 2 No. 3, pp. 97-101, Sept. 1998. DOI: http://dx.doi.org/10.1109/4235.735432