Abstract
This paper describes an architecture that begins with enough general knowledge to play any board game as a novice, and then shifts its decision-making emphasis to learned, game-specific, spatially-oriented heuristics. From its playing experience, it acquires game-specific knowledge about both patterns and spatial concepts. The latter are proceduralized as learned, spatially-oriented heuristics. These heuristics represent a new level of feature aggregation that effectively focuses the program's attention. While training against an external expert, the program integrates these heuristics robustly. After training it exhibits both a new emphasis on spatially-oriented play and the ability to respond to novel situations in a spatially-oriented manner. This significantly improves performance against a variety of opponents. In addition, we address the issue of context on pattern learning. The procedures described here move toward learning spatially-oriented heuristics for autonomous programs in other spatial domains.
Similar content being viewed by others
References
G. Biswas, S. Goldman, D. Fisher, B. Bhuva, & G. Glewwe. (1995). Assessing design activity in complex CMOS circuit design. In P. Nichols, S. Chipman, & R. Brennan (Eds.), Cognitively Diagnostic Assessment. Lawrence Erlbaum, Hillsdale, NJ.
A. Blum. (1997). Empirical support for Winnow and weighted-majority based algorithms: Results on a calendar scheuling domain. Machine Learning 26: 5–23.
S. Chatterjee & S. Chatterjee. (1987). On combining expert opinions. American Journal of Mathematical and Management Sciences 7(3/4): 271–295.
D. I. A. Cohen. (1972). The solution of a simple game. Mathematics Magazine 45(4): 213–216.
K. Crowley & R. S. Siegler. (1993). Flexible strategy use in young children's tic-tac-toe. Cognitive Science 17(4): 531–561.
M. J. Egenhofer & D.M. Mark. (1995). Naive geography, 95–8. National Center for Geographic Information and Analysis.
S. L. Epstein. (1992). Prior knowledge strengthens learning to control search in weak theory domains. International Journal of Intelligent Systems 7: 547–586.
S. L. Epstein. (1994a). For the right reasons: The FORR architecture for learning in a skill domain. Cognitive Science 18(3): 479–511.
S. L. Epstein. (1994b). Toward an ideal trainer. Machine Learning 15(3): 251–277.
S. L. Epstein, J. Gelfand, & J. Lesniak. (1996). Pattern-based learning and spatially-oriented concept formation with a multi-agent, decision-making expert. Computational Intelligence 12(1): 199–221.
R. Fine. (1989). The Ideas behind the Chess Openings. Random House, New York.
I. Gelfer. (1991). Positional Chess Handbook. Macmillan, New York.
E. Goldstein. (1989). Sensation and Perception (third ed.). Wadsworth, Belmont, CA.
Y. Ishida. (1991). All about Thickness. Ishi Press, San Jose, CA.
R. A. Jacobs. (1995). Methods for combining experts' probability assessments. Neural Computation 7: 867–888.
E. Kandel. (1991). Chapter 30: Perception of motion, depth, and form. In E. Kandel, J. Schwartz, & T. Jessel (Eds.), Principles of Neural Science, Elsevier, Amsterdam, pages 440–466.
B. Kuipers & T. Levitt. (1990). Navigation and mapping in large-scale space. In S. Chen (Ed.), Advances in Spatial Reasoning, Ablex, Norwood, NJ, pages 207–251.
R. Levinson & R. Snyder. (1991). Adaptive pattern-oriented chess. In Proceedings of the Eighth International Machine Learning Workshop, Morgan Kaufmann, San Mateo, pages 85–89.
N. Littlestone. (1988). Learning quickly when irrelevant attributes abound: A new linear-threshold algorithm. Machine Learning 2: 285–318.
H. Otake. (1992). Good shape. In Opening Theory Made Easy, Ishi Press, San Jose, CA, pages 62–111.
K. Penev & A. Requicha. (1997). Automatic fixture synthesis in 3D. In Proceedings of the IEEE Conference on Robotics and Automation, IEEE Press, Albuquerque, NM, pages 1713–1718.
M. J. Ratterman & S. L. Epstein. (1995). Skilled like a person: A comparison of human and computer game playing. In Proceedings of the Seventeenth Annual Conference of the Cognitive Science Society, Lawrence Erlbaum Associates, Pittsburgh, pages 709–714.
M. J. Ratterman & S. L. Epstein. Game-Playing Expertise in Young Children. In preparation.
L. Saitta, F. Neri, M. Bajo, J. Canas, S. Chaiklin, F. Esposito, D. Kayser, C. Nedellec, G. Sabah, A. Tiberghien, G. Vergnaud & S. Vosniadou (1995). Knowledge representation changes in humans and machines. In P. Reimenn, & H. Spada (Eds.), Learning in Humans and Machines, Elsevier Science, Oxford.
A. L. Samuel. (1963). Some studies in machine learning using the game of checkers. In E. A. Feigenbaum, & J. Feldman (Eds.), Computers and Thought. McGraw-Hill, New York, pages 71–105.
H. A. Simon. (1981). The Sciences of the Artificial (second ed.). MIT Press, Cambridge, MA.
A. Treisman. (1986). Features and objects in visual processing. Scientific American 255: 114–125.
B. Tversky. (1989). Parts, partonomies, and taxonomies. Developmental Psychology 25(6): 983–995.
P. Utgoff & D. Precup. (1997). Constructive function approximation. Technical Report 97–04. Department of Computer Science, University of Massachusetts.
J. Vandenbrande & A. Requicha. (1993). Spatial reasoning for the automatic recognition of machinable features in solid models. IEEE Transactions on Pattern Analysis and Machine Intelligence 15: 1269–1285.
D. Waco & Y. Kim. (1994). Reasoning for machining features using convex decomposition. Computer-Aided Design 26.
S. Zeki. (1993). A Vision of the Brain. Blackwell Scientific, Oxford.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Epstein, S.L., Gelfand, J. & Lock, E. Learning Game-Specific Spatially-Oriented Heuristics. Constraints 3, 239–253 (1998). https://doi.org/10.1023/A:1009734028965
Issue Date:
DOI: https://doi.org/10.1023/A:1009734028965