Machine Learning

, Volume 1, Issue 1, pp 11–46 | Cite as

Chunking in Soar: The anatomy of a general learning mechanism

  • John E. Laird
  • Paul S. Rosenbloom
  • Allen Newell
Article

Abstract

In this article we describe an approach to the construction of a general learning mechanism based on chunking in Soar. Chunking is a learning mechanism that acquires rules from goal-based experience. Soar is a general problem-solving architecture with a rule-based memory. In previous work we have demonstrated how the combination of chunking and Soar could acquire search-control knowledge (strategy acquisition) and operator implementation rules in both search-based puzzle tasks and knowledge-based expert-systems tasks. In this work we examine the anatomy of chunking in Soar and provide a new demonstration of its learning capabilities involving the acquisition and use of macro-operators.

Key words

learning from experience general learning mechanisms problem solving chunking production systems macro-operators transfer 

References

  1. Anderson, J.R. (1983). The architecture of cognition. Cambridge: Harvard University Press.Google Scholar
  2. Anderson, J.R. (1983). Knowledge compilation: The general learning mechanism. In R.S. Michalski, J.G. Carbonell, & T.M. Mitchell (Eds.). Proceedings of the 1983 Machine Learning Workshop. University of Illinois at Urbana-Champaign.Google Scholar
  3. Anzai, Y., & Simon, H.A. (1979). The theory of learning by doing. Psychological Review, 86, 124–140.CrossRefPubMedGoogle Scholar
  4. Brown, J.S., & VanLehn, K. (1980). Repair theory: A generative of bugs in procedural skills. Cognitive Science, 4, 379–426.CrossRefGoogle Scholar
  5. Carbonell, J.G., Michalski, R.S., & Mitchell, T.M. (1983). An overview of machine learning. In R.S. Michalski, J.G. Carbonell, T.M. Mitchell (Eds.). Machine learning: An artificial intelligence approach. Los Altos, CA: Morgan Kaufmann.Google Scholar
  6. Chase, W.G., & Simon, H.A. (1973). Perception in chess. Cognitive Psychology, 4 55–81.CrossRefGoogle Scholar
  7. Davis, R., & King, J. (1976). An overview of production systems. In E.W. Elcock & D. Michie (Ed.), Machine intelligence 8. New York: American Elsevier.Google Scholar
  8. DeJong, G. (1981). Generalizations based on explanations. Proceedings of the Seventh International Joint Conference on Artificial Intelligence (pp. 67–69). Vancouver, B.C., Canada: Morgan Kaufmann.Google Scholar
  9. Feigenbaum, E.A., & Feldman, J. (Eds.) (1963). Computers and thought. New York: McGraw-Hill.Google Scholar
  10. Fikes, R.E., Hart, P.E. & Nilsson, N.J. (1972). Learning and executing generalized robot plants. Artificial intelligence, 3, 251–288.CrossRefGoogle Scholar
  11. Forgy, C.L. (1981). OPS5 manual (Technical Report). Pittsburgh, PA: Computer Science Department, Carnegie-Mellon University.Google Scholar
  12. Forgy, C.L. (1982). Rete: A fast algorithm for the many pattern/many object pattern match problem. Artificial intelligence, 19, 17–37.CrossRefGoogle Scholar
  13. Forgy, C.L. (1984). The OPS83 Report (Tech. Rep. #84–133). Pittsburgh, PA: Computer Science Department, Carnegie-Mellon University.Google Scholar
  14. Hayes, J.R., & Simon, H.A. (1976). Understanding complex task instructions. In Klahr, D.(Ed.), Cognition and instruction. Hillsdale, NJ: Erlbaum.Google Scholar
  15. Korf, R.E. (1980). Toward a model of representation changes. Artificial intelligence, 14, 41–78.Google Scholar
  16. Korf, R.E. (1985). Macro-operators: A weak method for learning. Artificial intelligence, 26, 35–77.CrossRefGoogle Scholar
  17. Korf, R.E. (1985). Depth-first iterative-deepening: An optimal admissable tree search. Artificial intelligence, 27, 97–110.CrossRefMathSciNetGoogle Scholar
  18. Laird, J.E. (1984). Universal subgoaling. Doctoral dissertation, Computer Science Department, Carnegie-Mellon University, Pittsburgh, PA.Google Scholar
  19. Laird, J.E., & Newell, A. (1983). A universal weak method: Summary of results, Proceedings of the Eighth International Joint Conference on Artificial Intelligence (pp. 771–773). Karlsruhe, West Germany: Morgan Kaufmann.Google Scholar
  20. Laird, J.E., & Newell, A. (1983). A universal weak method (Tech. Rep. #83–141). Pittsburgh, PA: Computer Science Department, Carnegie-Mellon University.Google Scholar
  21. Laird, J.E., Newell, A., & Rosenbloom, P.S. (1985). Soar: An architecture for general intelligence. In preparation.Google Scholar
  22. Laird, J.E., Rosenbloom, P.S., & Newell, A. (1984). Towards chunking as a general learning mechanism. Proceedings of the National Conference on Artificial Intelligence (pp. 188–192). Austin, TX: Morgan Kaufmann.Google Scholar
  23. Langley, P. (1983). Learning Effective Search Heuristics. Proceedings of the Eighth International Joint Conference on Artificial Intelligence (pp. 419–425). Karlsruhe, West Germany: Morgan Kaufmann.Google Scholar
  24. Lenat, D. (1976). AM: An artificial intelligence approach to discovery in mathematics as heuristic search. Doctoral dissertation, Computer Science Department, Stanford University, Stanford, CA.Google Scholar
  25. Lenat, D.B. (1983). Eurisko: A program that learns new heuristics and domain concepts. Artificial intelligence, 21, 61–98.Google Scholar
  26. Lewis, C.H. (1978). Production system models of practice effects. Doctoral dissertation, University of Michigan, Ann Arbor, Michigan.Google Scholar
  27. Marsh, D. (1970). Memo functions, the graph traverser, and a simple control situation. In B. Meltzer & D. Michie (Eds.), Machine intelligence 5. New York: American Elsevier.Google Scholar
  28. McDermott, J. (1982). R1: A rule-based configurer of computer systems. Artificial intelligence, 19, 39–88.CrossRefGoogle Scholar
  29. Michie, D. (1968). ‘Memo’ functions and machine learning. Nature, 218, 19–22.Google Scholar
  30. Miller, G.A. (1956). The magic number seven, plus or minus two: Some limits on our capacity for processing information. Psychological Review, 63, 81–97.PubMedGoogle Scholar
  31. Mitchell, T.M. (1983). Learning and problem solving. Proceedings of the Eighth International Joint Conference on Artificial Intelligence (pp. 1139–1151). Karlsruhe, West Germany: Morgan Kaufmann.Google Scholar
  32. Mitchell, T.M., Keller, R.M., & Kedar-Cabelli, S.T. (1986). Explanation-based generalization: A unifying view. Machine learning, 1: 47–80.Google Scholar
  33. Neves, D.M., & Anderson, J.R. (1981). Knowledge compilation: Mechanisms for the automatization of cognitive skills. In Anderson, J.R. (Ed.), Cognitive skills and their acquisition. Hillsdale, NJ: Erlbaum.Google Scholar
  34. Newell, A. (1973). Production systems: Models of control structures. In Chase, W. (Ed.). Visual information processing. New York: Academic.Google Scholar
  35. Newell, A. (1980). Reasoning, problem solving and decision processes: The problem space as a fundamental category. In R. Nickerson (Ed.), Attention and performance VIII. Hillsdale, N.J.: Erlbaum. (Also available as CMU CSD Technical Report, Aug 79).Google Scholar
  36. Newell, A., & Rosenbloom, P.S. (1981). Mechanisms of skill acquisition and the law of practice. In J.R. Anderson (Ed.), Cognitive skills and their acquisition. Hillsdale, NJ: Erlbaum. (Also available as Carnegie-Mellon University Computer Science Tech. Rep. #80–145).Google Scholar
  37. Nilsson, N. (1980). Principles of artificial intelligence, Palo Alto, CA: Tioga.Google Scholar
  38. Rendell, L.A. (1983). A new basis for state-space learning systems and a successful implementation. Artificial intelligence, 20, 369–392.CrossRefGoogle Scholar
  39. Rich, E. (1983). Artificial intelligence. New York: McGraw-Hill.Google Scholar
  40. Rosenbloom, P.S. (1983). The chunking of goal hierarchies: A model of practice and stimulus-response compatibility. Doctoral dissertation, Carnegie-Mellon University, Pittsburgh, PA. (Available as Carnegie-Mellon University Computer Science Tech. Rep. #83-148).Google Scholar
  41. Rosenbloom, P.S., & Newell, A. (1986). The chunking of goal hierarchies: A generalized model of practice. In R.S. Michalski, J.G. Carbonell, & T.M. Mitchell (Eds.), Machine Learning: An Artificial Intelligence Approach, Volume II. Los Altos, CA: Morgan Kaufmann Publishers, Inc. In press (Also available in Proceedings of the Second International Machine Learning Workshop, Urbana: 1983).Google Scholar
  42. Rosenbloom, P.S., Laird, J.E., McDermott, J., Newell, A., & Orciuch, E. (1985). R1-Soar: An experiment in knowledge-intensive programming in a problem-solving architecture. IEEE Transactions on Pattern Analysis and Machine Intelligence, 7, 561–569. (Also available in Proceedings of the IEEE Workshop on Principles of Knowledge-Based Systems, Denver: IEEE Computer Society, 1984, and as part of Carnegie-Mellon University Computer Science Tech. Rep. #85–110).Google Scholar
  43. Smith, R.G., Mitchell, T.M., Chestek, R.A., & Buchanan, B.G. (1977). A model for learning systems. Proceedings of the Fifth International Joint Conference on Artificial Intelligence. (pp. 338–343). Cambridge, Mass.: Morgan Kaufmann.Google Scholar
  44. Sussman, G.J. (1977). A computer model of skill acquisition. New York: Elsevier.Google Scholar
  45. Utgoff, P.E. (1984). Shift of bias for inductive concept learning. Doctoral dissertation, Rutgers University, New Brunswick, NJ.Google Scholar
  46. van de Brug, A., Rosenbloom, P.S., & Newell, A. (1985). Some experiments with R1-Soar (Tech. Rep.). Computer Science Department, Carnegie-Mellon University, Pittsburgh, PA. In preparation.Google Scholar
  47. Waterman, D.A. (1975). Adaptive production systems. Proceedings of the Fourth International Joint Conference on Artificial Intelligence (pp. 296–303). Tbilisi, USSR: Morgan Kaufmann.Google Scholar

Copyright information

© Kluwer Academic Publishers 1986

Authors and Affiliations

  • John E. Laird
    • 1
  • Paul S. Rosenbloom
    • 2
    • 3
  • Allen Newell
    • 4
  1. 1.Intelligent Systems LaboratoryXerox Palo Alto Research CenterPalo AltoU.S.A.
  2. 2.Department of Computer ScienceStanford UniversityStanfordU.S.A.
  3. 3.Department of PsychologyStanford UniversityStanfordU.S.A.
  4. 4.Department of Computer ScienceCarnegie-Mellon UniversityPittsburghU.S.A.

Personalised recommendations