A Subsymbolic and Symbolic Model for Learning Sequential Decision Tasks

  • Ron Sun
  • Todd Peterson
Part of the Studies in Fuzziness and Soft Computing book series (STUDFUZZ, volume 59)


For dealing with reactive sequential decision tasks, a learning model Clarion was developed, which is a hybrid connectionist model consisting of both localist (symbolic) and distributed representations, based on the two-level approach proposed in Sun (1995). The model learns and utilizes procedural and declarative knowledge, tapping into the synergy of the two types of processes. It unifies neural, reinforcement, and symbolic methods to perform on-line, bottom-up learning (from subsymbolic to symbolic knowledge). Experiments in various situations shed light on the working of the model. Its theoretical implications in terms of symbol grounding are also discussed.


Symbolic Representation Bottom Level Inductive Logic Programming Declarative Knowledge Navigation Task 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. J. R. Anderson, (1983). The Architecture of Cognition, Harvard University Press, Cambridge, MAGoogle Scholar
  2. D. Bertsekas and J. Tsitsiklis, (1996). Neuro-Dynamic Programming. Athena Scientific, Belmont, MA.Google Scholar
  3. L. Breiuran, (1996). Bagging predictors. Machine Learning, Vol. 24, No. 2, pp. 123–140.Google Scholar
  4. H. Dreyfus and S. Dreyfus, (1987). Mind Over Machine. The Free Press, New York, NY.Google Scholar
  5. D. Fisher, (1987). Knowledge acquisition via incremental conceptual clustering. Machine Learning. 2, 139–172.Google Scholar
  6. P. Fitts and M. Posner, (1967). Human Performance. Brooks/Cole, Monterey, CA.Google Scholar
  7. S. Hamad, (1990). The symbol grounding problem. Physica D, 42, 335–346.CrossRefGoogle Scholar
  8. H. Hirsh, (1994). Generalizing version spaces. Machine Learning, 17, 5–46.MATHGoogle Scholar
  9. W. James, (1890). The Principles of Psychology. Dover, New York.CrossRefGoogle Scholar
  10. L. Kaelbling, M. Littman, and A. Moore, (1996). Reinforcement learning: A survey. Journal of Artificial Intelligence Research, 4, 237–285.Google Scholar
  11. F. Keil, (1989). Concepts, Kinds, and Cognitive Development. MIT Press. Cambridge, MA.Google Scholar
  12. N. Lavrac and S. Dzeroski, (1994). Inductive Logic Programming. Lllis Horword, New York.Google Scholar
  13. L. Lin, (1992). Self-improving reactive agents based on reinforcement learning, plan-ning, and teaching. Machine Learning. Vol. 8, pp. 293–321.Google Scholar
  14. R. Michalski, (1983). A theory and methodology of inductive learning. Artificial Intelligence. Vol.20, pp.111–161. Google Scholar
  15. T. Mitchell, (1982). Generalization as search. Artificial Intelligence, 18, 203–226.MathSciNetCrossRefGoogle Scholar
  16. G. Monohan, (1982). A survey of partially observable Markov decision processes: theory, models, and algorithms. Management Science, 28 (1), 1–16.MathSciNetCrossRefGoogle Scholar
  17. R. Quinlan, (1986). Inductive learning of decision trees. Machine Learning. 1, 81–106.Google Scholar
  18. P. Rosenbloom, J. Laird, A. Newell, and R. McCarl, (1991). A preliminary analysis of the SOAR architecture as a basis for general intelligence. Artificial Intelligence. 47 (1–3), 289–325.MathSciNetCrossRefGoogle Scholar
  19. Technology. 213–225. U. of New Hampshire, Durham.Google Scholar
  20. P. Smolensky, (1988). On the proper treatment of connectionism. Behavioral and Brain Sciences, 11 (1): 1–74.MathSciNetCrossRefGoogle Scholar
  21. R. Sun, (1992). On variable binding in connectionist networks, Connection Science, Vol. 4, No. 2, pp. 93–124.CrossRefGoogle Scholar
  22. R. Sun, (1994). Integrating. Rules and Connectionism for Robust Commonsense Reasoning. John Wiley and Sons, New York, NY.Google Scholar
  23. R. Sun, (1995). Robust reasoning: integrating rule-based and similarity-based reasoning. Artificial Intelligence. 75, 2. 241–296.CrossRefGoogle Scholar
  24. R. Sun, (1997). Learning, action, and consciousness: a hybrid approach towards modeling consciousness. Neural Networks, special issue on consciousness. 10 (7), pp. 1317–1331.Google Scholar
  25. R. Sun and F. Alexandre, (eds.) (1997). Connectionist Symbolic Integration. Lawrence Erlbaum Associates, Hillsdale, NJ.Google Scholar
  26. R. Sun and T. Peterson, (1995). A hybrid learning model of reactive sequential decision making. In: R. Sun and F. Alexandre, (eds.) The Working Notes of The IJCA I Workshop on Connectionist-Symbolic Integration.Google Scholar
  27. R. Sun and T. Peterson, (1998). Some experiments with a hybrid model for learning sequential decision making. Information Sciences. Vol. 111, pp. 83–107.CrossRefGoogle Scholar
  28. R. Sutton, (1990). Integrated architectures for learning, planning, and reacting based on approximating dynamic programming. Proc.of Seventh International Conference on Machine Learning. Morgan Kaufmann. San Mateo, CA.Google Scholar
  29. T. Tesauro, (1992). Practical issues in temporal difference learning. Machine Learning. Vol. 8, 257–277.MATHGoogle Scholar
  30. G. Towell and J. Shavlik, (1993). Extracting Refined Rules from Knowledge-Based Neural Networks, Machine Learning. 13 (1), 71–101.Google Scholar
  31. P. Utgoff (1989). Incremental induction of decision trees. Machine Learning. Vol. 4, 161–186.CrossRefGoogle Scholar
  32. C. Watkins, (1989). Learning with Delayed Rewards. Ph.D Thesis, Cambridge University, Cambridge, UK.Google Scholar
  33. D. Willingham, M. Nissen, and P. Bullemer, (1989). On the development of procedural knowledge. Journal of Experimental Psychology: Learning, Memory, and Cognition. 15, 1047–1060.Google Scholar

Copyright information

© Physica-Verlag Heidelberg 2001

Authors and Affiliations

  • Ron Sun
    • 1
    • 2
  • Todd Peterson
    • 1
    • 2
  1. 1.NEC Research InstitutePrincetonUSA
  2. 2.The University of AlabamaTuscaloosaUSA

Personalised recommendations