코호넨 신경망을 이용 바둑 사활문제를 풀기 위한 후보 첫 수들

Candidate First Moves for Solving Life-and-Death Problems in the Game of Go, using Kohonen Neural Network

  • Lee, Byung-Doo (Dept. of Information and Communication Engineering, Sungkyul University) ;
  • Keum, Young-Wook (Dept. of Computer Engineering, Sungkyul University)
  • 발행 : 2009.02.20

초록

바둑에 있어 사활문제는 컴퓨터 바둑을 구현하기 위해 반드시 극복해야 하는 기본적인 문제이다. 사활문제와 같은 국부적인 바둑 문제를 해결하기 위하여 고려해야 될 중요한 사항은 게임 트리의 엄청난 분기수와 그 깊이를 어떻게 처리하느냐이다. 본 논문에서 수행된 실험의 기본 착상은 둘러싸인 돌들을 죽이기 위해 인식된 첫 수들을 찾아내는 인간의 습성을 모방한 것이다. 바둑에 있어, 유사한 사활문제(패턴)들은 자주 유사한 해들을 갖는다. 유사한 패턴을 분류 하기 위하여 코호넨 신경망(KNN)을 기반으로 한 군집화를 수행하였으며, 실험 결과는 고무적이며 사활문제를 풀기 위해 신경망으로 통제 학습을 사용하는 패턴 일치와 경쟁할 수 있음을 알아냈다.

In the game of Go, the life-and-death problem is a fundamental problem to be definitely overcome when implementing a computer Go program. To solve local Go problems such as life-and-death problems, an important consideration is how to tackle the game tree's huge branching factor and its depth. The basic idea of the experiment conducted in this article is that we modelled the human behavior to get the recognized first moves to kill the surrounded group. In the game of Go, similar life-and-death problems(patterns) often have similar solutions. To categorize similar patterns, we implemented Kohonen Neural Network(KNN) based clustering and found that the experimental result is promising and thus can compete with a pattern matching method, that uses supervised learning with a neural network, for solving life-and-death problems.

키워드

참고문헌

  1. J. Burmeister, “Studies in Human and Computer Go: Assessing the Game of Go as a Research Domain for Cognitive Science”, PhD thesis, University of Queensland, Australia, 1995.
  2. N. Sasaki, Y. Sawada, J. Yoshimura, “A Neural Network Program of Tsume-Go”, Computer and Games 1998, pp. 167-182, Springer-Verlag, Germany, 1999.
  3. T. Kohonen, “The Self-Organizing Map”, Springer, Germany, 2nd Edition, 1997.
  4. M. Harada, “Tsumego - Life and Death Problem of Go”, http://www.hitachi.co.jp/Sp/tsumego/ indexe.html.
  5. T. Thireou, L. Strauss, G. Kontaxakis, S. Pavlopoulos, A. Santos, “Principal Component Analysis in Dynamic Position Emission Tomography”, IEA/AIE 1998, vol. 2, pp. 612-618, 1998.
  6. S. Raychaudhuri, J. M. Stuart, R. B. Altman, “Principal Components Analysis to Summarize Microarray Experiments: Application to Sporulation Time Series”, Pacific Symposium on Biocomputing 2000, pp. 452-463, USA, 2000.
  7. A. Kishimoto, “Transposition table driven scheduling for two-players game”, Master's thesis, University of Alberta, Canada, 2002.
  8. H. Liu, H. Motoda, Feature Extraction, Construction and Selection: A Data Mining Perspective, Kluwer Academic Publishers, USA, 1998.
  9. J. Mao, A. K. Jain, “A Self-Organizing Network for Hyperellipsoidal Clustering (HEC)”, IEEE Transactions on Neural Networks, 7(1):16-29, 1996. https://doi.org/10.1109/72.478389
  10. K. Yeung, W. Ruzzo, “An Empirical Study on Principal Component Analysis for Clustering Gene Expression Data”, Bioinformatics, 17(9): 763-774, 2001. https://doi.org/10.1093/bioinformatics/17.9.763
  11. K. Das, W. Perrizo, Q. Ding, “Using Neural Networks for Clustering on RSI Data and Related Spatial Data”, ISCA International Conference Reuse and Integration, pp. 111-116, USA, 2000.
  12. Z. Zupan, Z. Gasteiger, Neural Networks in Chemistry and Drug Design, Wiley-VCH, 1999.
  13. S. Lawrence, C. L. Giles, A. C. Tsoi, A. D. Back, “Face Recognition: A Hybrid Neural Network Approach”, Technical report UMIACSTR-96-16 and CS-TR-3608, Institute for Advanced Computer Studies University of Maryland, USA, 1996.
  14. T. Reutterer, M. Natter, “Segmentation Based Competitive Analysis with MULTICLUS and Topology Representing Networks”, Computers and Operations Research 2000, 27(11-12): 1227-1247, 2000. https://doi.org/10.1016/S0305-0548(99)00147-1
  15. T. Kohonen, “The Self-Organizing Map”, IEEE, vol. 78, pp. 1464-1480, 1990. https://doi.org/10.1109/5.58325