Predictive and Descriptive Approaches to Learning Game Rules from Vision Data

  • Paulo Santos
  • Simon Colton
  • Derek Magee
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4140)


Systems able to learn from visual observations have a great deal of potential for autonomous robotics, scientific discovery, and many other fields as the necessity to generalise from visual observation (from a quotidian scene or from the results of a scientific enquiry) is inherent in various domains. We describe an application to learning rules of a dice game using data from a vision system observing the game being played. In this paper, we experimented with two broad approaches: (i) a predictive learning approach with the Progol system, where explicit concept learning problems are posed and solved, and (ii) a descriptive learning approach with the HR system, where a general theory is formed with no specific problem solving task in mind and rules are extracted from the theory.


Vision System Production Rule Vision Data Inductive Logic Programming Descriptive Approach 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Colton, S.: Automated Theory Formation in Pure Mathematics. Springer, Heidelberg (2002)Google Scholar
  2. 2.
    Colton, S., Muggleton, S.: ILP for mathematical discovery. In: Proceedings of the 13th International Conference on Inductive Logic Programming (2003)Google Scholar
  3. 3.
    Colton, S., Bundy, A., Walsh, T.: On the notion of interestingness in automated mathematical discovery. International Journal of Human Computer Studies 53(3), 351–375 (2000)MATHCrossRefGoogle Scholar
  4. 4.
    de Raedt, L., Dehaspe, L.: Clausal discovery. Machine Learning 26, 99–146 (1997)MATHCrossRefGoogle Scholar
  5. 5.
    Dehaspe, L., Toivonen, H.: Discovery of frequent datalog patterns. Data Mining and Knowledge Discovery 3(1), 7–36 (1999)CrossRefGoogle Scholar
  6. 6.
    King, R., Whelan, K., Jones, F., Reiser, P., Bryant, C., Muggleton, S., Kell, D., Oliver, S.: Functional genomic hypothesis generation and experimentation by a robot scientist. Nature 427, 247–252 (2004)CrossRefGoogle Scholar
  7. 7.
    Magee, D.R.: Tracking multiple vehicles using foreground, background and motion models. Image and Vision Computing 20(8), 581–594 (2004)CrossRefGoogle Scholar
  8. 8.
    Magee, D.R., Needham, C.J., Santos, P.E., Rao, S.: Inducing the focus of attention by observing patterns in spac. In: IJCAI Workshop on Modelling Others from Observations (MOO 2005), pp. 47–52 (2005)Google Scholar
  9. 9.
    McCune, W.: The OTTER user’s guide. Technical Report ANL/90/9, Argonne National Laboratories (1990)Google Scholar
  10. 10.
    Muggleton, S.: Learning from positive data. In: International Conference on Inductive Logic Programming (1996)Google Scholar
  11. 11.
    Muggleton, S.: CProgol4.4: a tutorial introduction. In: Relational Data Mining, pp. 160–188. Springer, Heidelberg (2001)Google Scholar
  12. 12.
    Needham, C., Santos, P., Magee, D., Devin, V., Hogg, D., Cohn, A.: Protocols from perceptual observations. Artificial Intelligence Journal 167, 103–136 (2005)CrossRefGoogle Scholar
  13. 13.
    Quinlan, J., Cameron-Jones, R.: FOIL: A midterm report. In: Brazdil, P.B. (ed.) ECML 1993. LNCS (LNAI), vol. 667. Springer, Heidelberg (1993)Google Scholar
  14. 14.
    Stehl, A., Ghosh, J.: Cluster ensembles – a knowledge reuse framework for combining multiple partitions. Journal of Machine Learning Research 3, 583–617 (2002)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Paulo Santos
    • 1
  • Simon Colton
    • 2
  • Derek Magee
    • 3
  1. 1.Department of Electrical EngineeringCentro Universitário da FEISão PauloBrazil
  2. 2.Department of ComputingImperial CollegeLondonUK
  3. 3.School of ComputingLeeds UniversityLeedsUK

Personalised recommendations