Journal of KIISE:Software and Applications (한국정보과학회논문지:소프트웨어및응용)

Korean Institute of Information Scientists and Engineers (한국정보과학회)
권호 표지

New Usage of SOM for Genetic Algorithm

유전 알고리즘에서의 자기 조직화 신경망의 활용

  • 김정환 (서울대학교 컴퓨터공학부) ;
  • 문병로 (서울대학교 컴퓨터공학부)
  • Published : 2006.04.01
Download PDF
⟨ Previous Next ⟩

Abstract

Self-Organizing Map (SOM) is an unsupervised learning neural network and it is used for preserving the structural relationships in the data without prior knowledge. SOM has been applied in the study of complex problems such as vector quantization, combinatorial optimization, and pattern recognition. This paper proposes a new usage of SOM as a tool for schema transformation hoping to achieve more efficient genetic process. Every offspring is transformed into an isomorphic neural network with more desirable shape for genetic search. This helps genes with strong epistasis to stay close together in the chromosome. Experimental results showed considerable improvement over previous results.

자기 조직화 신경망 (SOM: Self-Organizing Map)은 자율 학습 신경망으로 사전 지식이 존재하지 않는 자료에 존재하는 구조적 관계성을 보전하는데 이용된다. 자기 조직화 신경망은 벡터 양자화, 조합 최적화, 패턴 인식과 같은 복잡한 문제 해결을 위한 연구에 많이 이용되어 왔다. 이 논문에서는 좀더 효율적인 유전 알고리즘을 얻기 위한 스키마 변환 도구로서 자기 조직화 신경망을 이용하는 새로운 사용법에 대해서 제안한다. 즉, 각 자식해는 탐색 공간에서 좀더 바람직한 모양을 가지는 동질의 인공 신경망으로 변환된다. 이 변환으로 인해 강한 상위(epistasis)를 가지는 유전자들은 염색체 상에서 서로 인접하게 되는 것이다. 실험 결과는 기존 결과에 비해서 주목할만한 성능 개선이 있음을 보여준다.

Keywords

References

  1. S. Haykin. Neural Networks: A Comprehensive Foundation. Pretice Hall, 1999
  2. A. F. James and M. S. David. Neural Networks, Algorithms, Applications, and Programming Techniques. Addison Wesley, 1994
  3. A. Grauel and F. Berk. Mapping of dynamical systems by recurrent neural networks in an evolutionary algorithm approach. In European Congress on Intelligent Techniques and Soft Computing, volume 1, pages 470-476, 1998
  4. R. Jeff and V. B. Ciesielski. An evolutionary approach to training feedforward and recurrent neural networks. In International Conference on Knowledge-Based Intelligent Electronics Systems, pages 596-602, 1998 https://doi.org/10.1109/KES.1998.726028
  5. V. Patridis, E. Paterakis, and A. Kehagias. A hybrid neural-genetic multimodel parameter estimation algorithm. IEEE Transactions on Neural Networks, 9(5):862-876, 1998 https://doi.org/10.1109/72.712158
  6. J. D. Schaffer, D. Whitely, and L. J. Eshelman. Combinations of genetic algorithms and neural networks: A survey of the state of the art. In International Workshop on Combinations of Genetic Algorithms and Neural Networks, pages 1-37, 1992 https://doi.org/10.1109/COGANN.1992.273950
  7. C. A. Anderson, K. F. Jones, and J. Ryan. A two-dimensional genetic algorithm for the Ising problem. Complex Systems, 5:327-333, 1991
  8. J. P. Cohoon and W. Paris. Genetic placement. In IEEE International Conference on Computer-Aided Design, pages 422-425, 1986
  9. T. N. Bui and B. R. Moon. On multi-dimensional encoding/crossover. In International Conference of Genetic Algorithms, pages 49-55, 1995
  10. A. B. Kahng and B. R. Moon. Toward more powerful recombinations. In International Conference on Genetic Algorithms, pages 96-103, 1995
  11. J. H. Kim and B. R. Moon. Neuron reordering for better neuro-genetic hybrids. In Genetic and Evolutionary Computation Conference, pages 407-414, 2002
  12. B. R. Moon and H. N. Kim. Effective genetic coding with a two-dimensional embedding heuristic. International Journal of Knowledge- Based Intelligent Engineering Systems, 3(2):113-120, 1999
  13. B. R. Moon, Y. S. Lee and C. K. Kim GEORG: VLSI circuit partitioner with a new genetic algorithm framework. Journal of Intelligent Manufacturing, 9(5):401-412, 1998 https://doi.org/10.1023/A:1008812214143
  14. J. L. Elman. Finding structure in time. Cognitive Science, 14:179-211, 1990 https://doi.org/10.1207/s15516709cog1402_1
  15. T. Kohonen, S. Kaski, K. Lagus, J. Salojarvi, J.Honkela, V, Paatero, and A. Saarela. Self organization of a massive document collection. IEEE Transactions on Neural Networks, 11(3):574-585, 2000 https://doi.org/10.1109/72.846729
  16. J. Bagley. The Behavior of Adaptation Systems Which Employ Genetic and Correlation Algorithms. PhD thesis, University of Michigan, Ann Arbor, MI, 1967
  17. T. N. Bui and B. R. Moon. Hyperplane synthesis for genetic algorithms. In International Conference of Genetic Algorithms, pages 102-109, 1993
  18. T. N. Bui and B. R. Moon. GRCA: A hybrid genetic algorithm for circuit ratio-cut partitioning. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 17(3):193-204, 1998 https://doi.org/10.1109/43.700718
  19. D. E. Rumelhart, G. E. Hinton, and R. J. Williams. Learning internal representations by error propagation, In D. E. Rumelhart and J. L. McCleland, editors, Parallel Distributed Processing, volume 1, chapter 8. MIT Press, Cambridge, MA, 1986
  20. D. Whitley and J. Kauth, Genitor: A different genetic algorithm. In Rocky Mountain Conference on Artificial Intelligence, pages 118-130, 1988
  21. D. Cavicchio, Adaptive search using simulated evolution. PhD thesis, Univ. of Michigan, Ann Arbor, Mich., 1970, Unpublished
  22. T. N. Bui and B. R. Moon. Genetic algorithm and graph partitioning. IEEE Transactions On Computers, 45(7):841-855, 1996 https://doi.org/10.1109/12.508322
  23. C. T. Lin and C. P. Jou. Controlling chaos by GA-based reinforcement learning neural network. IEEE Transactions on Neural Networks, 10(4): 846-869, 1999 https://doi.org/10.1109/72.774236