Abstract
Go is a difficult game to make a computer program because of the space complexity. Therefore, it is important to explore another approach that does not rely on search algorithms only. In this paper, we focus on tsume-go problems (local Go problems) that have a unique solution. A three-layer neural network program has been developed to find a solution at a given position of tsume-go problems, where the attacker is to kill the defender’s territory on a 9 × 9 board. The network consists of 162 neurons for the input layer, 300 neurons for the middle layer, and 81 neurons for the output layer. We let the network learn the current stone patterns and, hence, process a direct answer. The network learns 2000 patterns of tsume-go by the back-propagation method. Within 500 repeats, the network learns 2000 patterns correctly. We tested the network ability: the top three selected moves contain about 60% correct answers, and the top five, about 70% for unknown problems at 500 repeats of learning. We compare the rate of correct answers by the network with that of human players who replied in a few seconds only. The ability of the network is roughly equivalent to 1-dan strength of human player. Application of neural networks for a computer program of tsume-go (and also Go) combined with a pattern classifier might provide a prospective approach to create a strong Go-playing program.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
D.E. Rumelhart, G.E. Hinton, and R.J. Williams. Learning representations by backpropagating errors. Nature, vol. 323, pp. 533–536, 1986.
M. Enzenberger. The Integration of A Priori Knowledge into a Go Playing Neural Network. available from http://www.cip.physik.uni-muenchen.de/enz/NeuroGo/NeuroGo.html, 1996.
D. Fotland. Knowledge Representation in The Many Faces of Go. available from Go Archive Site as Go/comp/mfg.Z, 1993.
N. Richards, D. Moriarty, and R. Miikkulainen. Evolving Neural Networks to Play Go. Proceedings of the 7th International Conference on Genetic Algorithms, East Lansing, MI, 1997.
T. Wolf. About problems in generalizing a tsumego program to open positions. Proceedings of the 3rd Game Programming Workshop, Hakone, pp.20–26, 1996.
A. Tozawa. Tsume-go for 9-kyu to 1-kyu, (in Japanese). Seibidou Shuppan, ISBN4-415-04423-9, 1992.
T. Kojima, A. Yoshikawa, K. Ueda, and S. Nagano. Toward Building a Model of the Go Champion: Fusion between Cognitive Science and Artificial Intelligence, (in Japanese). Proceedings of 1997 Japanese Cognitive Science Society Winter Symposium “Game and Cognitive Science”, pp.25–31, 1997.
Y. Ishida. Fundamental tsume-go 100 Problems, (in Japanese). Nihon Bungeisha, ISBN4-537-01223-4, 1997.
Y. Ishida. Tsume-go 100 Problems to challenge 3-dan, (in Japanese). Tsuchiya Shoten, ISBN4-8069-1413-4, 1989.
Y. Ishida. Tsume-go 100 Problems to attain 5-dan, (in Japanese). Tsuchiya Shoten, ISBN4-8069-1414-4, 1989.
N. Sasaki. The Neural Network Programs for Games, (in Japanese). Ph.D. thesis, Graduate School of Information Science, Tohoku University, 1998.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1999 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Sasaki, N., Sawada, Y., Yoshimura, J. (1999). A Neural Network Program of Tsume-Go. In: van den Herik, H.J., Iida, H. (eds) Computers and Games. CG 1998. Lecture Notes in Computer Science, vol 1558. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-48957-6_10
Download citation
DOI: https://doi.org/10.1007/3-540-48957-6_10
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-65766-8
Online ISBN: 978-3-540-48957-3
eBook Packages: Springer Book Archive