A Genetic Algorithm with Local Competing

지역적으로 경정하는 유전자 알고리즘

  • 강태원 (강릉대학교 컴퓨터공학과)
  • Published : 2002.06.01

Abstract

On the whole, the simple GAs with just one population set is effective in finding one optimal solution. However, many real world problems have a lot of optimal solutions, and often it is important to find all of them. In this paper, we propose a GA that has a population set containing multiple optimal solutions. In the proposed GA, each of the individuals in population set has its own geological neighbors, and they exchange their genes globally as well as compete with others locally. The paper then evaluates the proposed GA along with many multimodal problems including a 30bit, order-six bipolar-deceptive function. Finally, we present some improvement directions of the proposed GA.

한 개의 모집단으로 구성되는 단순 유전자 알고리즘은 일반적으로 하나의 최적해를 찾는 경우에만 효과적이다. 그러나, 많은 문제들은 여러 개의 최적해를 가질 수 있으며, 그것들 모두를 찾는 것이 중요한 경우가 많다. 이 논문에서는 모집단 내 개체들에 지리적인 이웃의 개념을 부여하여, 각 객체들이 지역적으로 경쟁하면서도 전역적으로 유전자를 교환할 수 있도록 하여, 하나의 모집단이 여러 개의 최적해를 포함하도록 하는 유전자 알고리즘을 제안한다. 또한, 30비트, 6차 바이폴라-디셉티브 함수(bipolar-deceptive function)를 비롯한 여러 개의 최적해를 갖는 다양한 문제들에 적용하여 성능을 평가한다. 마지막으로 제안한 알고리즘에 대한 몇 가지 개선 방향을 제시하였다.

Keywords

References

  1. Z. Michalewicz, 'Genetic Algorithm+Data Structure=Evolutionary Programs,' Springer, pp. 169-172, 1999
  2. Georges Harik, Finding Multiple Solutions In Problems Of Bounded Difficulty, ⅢiGAL Report No. 94002, University of Illionis, The Illionis Genetic Algorithms Laboratory, 1994
  3. George Harik, Finding Multimodal Solutions Using Restricted Tourmament Selection, University of Illinois, The Illionis Genetic Algorithms Laboratory, 1994
  4. Poundstone, The Recursive Universe, New York: Morrow, 1985
  5. Langton CG, Artificial Life, Addision-Wesley, 1989
  6. Gardner Martin, The fantastic combinations fo John Conway's new solitaire game 'life,' Scientific American, pp. 120-123, October 1970
  7. Wolfram, S., Cellular automata and complexity, Reading, Mass. Addision-Wesley, 1994
  8. Basley, et. all, A sequential Niche Technique for Multimodal Function Optimization, Evolutionary Computation, Vol.1, No.2, pp. 101-125, 1993 https://doi.org/10.1162/evco.1993.1.2.101
  9. Mahfoud, Niching Methods for Genetic Algorithms, IlliGAL Report No. 95001, University of Illinois, The Illinois Genetic Algorithms Laboratory, 1995
  10. De Jong, An Analysis of the Behavior of a Class of Genetic Adaptive Systems, (Doctoral dissertation, University of Michgan), Dissertation Abstract International, 36(10), 5140B, (University Microfilms No 76-9381), 1975
  11. Mahfoud, A Comparison of Parallel and Sequential Niching Methods, Proc. Sixth International Conference on Genetic Algorithms, pp. 136-143, 1995
  12. Goldberg, Richardson, Genetic Algorithms with Sharing for Multimodal Function Optimization, Proc. 2nd International Conference on Genetic Algorithms, pp. 41-49, 1987
  13. Spears, Simple Subpopulations Schemes, Proc. 3rd Annual Conference on Evolutionary Programming, San Diego, CA, Word Scientific, pp. 367-384, 1994
  14. Erick Cantu-Paz, A Summary of Research on Parallel Genetic Algorithms, IlliGAL Report No.95007, University of Illinois, The Illinois Genetic Algorithms Laboratory, July 1995
  15. Prasanna V. Parthasarathy, David E. Goldberg and Scott A. Burns, Tracking Multimodal Problems in Hybrid Genetic Algorithms, University of Illinois, The Illinois Genetic Algorithms Laboratory, 2001
  16. Hancock, Selection Methods for Evolutionary Algorithms, in Practical Handbook of Genetic Algorithms, CRC Press, Vol. 2, pp. 67-92, 1995
  17. Pawlowsky, Crossover Operator, in Practical Handbook of Genetic Algorithms, CRC Press, Vol. 1, pp. 101-114, 1995
  18. Paul Rendell, Turning Machine implemented in Conway's Game Life, http://www.rendell.uk.co/gol/tm/htm, April 2000
  19. David E. Goldberg, Kalyanmoy Deb, Jeffrey Horn, Massive Multimodality, Deception, and Genetic Algorithms, Parallel Problem Solving from Nature, 2, North-Holland, pp. 37-46, 1992