Integrating an explanation-based learning mechanism into a general problem-solver

  • F. Zerr
  • J. G. Ganascia
Part 1: Constructive Induction And Multi-Strategy Approaches
Part of the Lecture Notes in Computer Science book series (LNCS, volume 482)


In this paper, we study the problem of integrating an Explanation-Based Learning mechanism into a general and industrial problem solver. During our investigations into SOAR we discovered a number of weaknesses related to its architecture and learning mechanism, known as Chunking. Using general concepts on which SOAR is based, we define a new learning system based, on the one hand, on a general and industrial problem solver, and, on the other hand, on an efficient learning mechanism known as EBG (Explanation-Based Generalization). Due to the fact that EBG sometimes learns production rules which are too general, we have introduced the possibility of restricting the generality of learned rules, in order to improve significantly the performances of industrial applications.

We have tested this system on an industrial application at Thomson. In this application, we had to find correct trajectories for planes through a network of valleys. This type of problem is complex and the search combinational, as the valleys were not properly interconnected. However, learning 327 production rules made the resolution 78 times faster and the system sometimes even found better solutions.


Explanation-based learning EBG Chunking problem solving production systems 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    W. Chehire, A. Combastel, D. Chouvet, J.Y. Quemeneur, F. Zerr. ”CIME: Une approche cohérente pour développer, intégrer et optimiser des modules experts dans un milieu opérationnel” in Avignon 1989Google Scholar
  2. [2]
    W. Chehire ”KIRK: Un environnement de développement de Systèmes Experts” in Avignon 1988Google Scholar
  3. [3]
    B. Clayton ”ART: Automated Reasoning Tool” Inference Corporation 1985Google Scholar
  4. [4]
    G. Dejong, R. Mooney ”Explanation-Based Learning: An alternative view” in Machine Learning 1: 145–176, 19. 1986Google Scholar
  5. [5]
    C. L. Forgy ”RETE: A fast algorithm for many pattern/many object pattern match problem.” In Artificial Intelligence 19:17–37, 1982CrossRefGoogle Scholar
  6. [6]
    J. Laird, A. Newell, P.S. Rosenbloom ”SOAR: An architecture for general intelligence” In Artificial Intelligence 33, 1–64, 1987CrossRefGoogle Scholar
  7. [7]
    J. Laird, P.S. Rosenbloom, A. Newell ”Chunking in SOAR: The anatomy of a general learning mechanism” in Machine Learning 1, 1986Google Scholar
  8. [8]
    J. Laird ”The SOAR Casebook” SOAR Project Papers CMU. 1986Google Scholar
  9. [9]
    T. Mitchell, R. Keller, S. Kedar-cabelli ”Explanation-Based Generalization: A unifying view” in Machine Learning 1, 1986Google Scholar
  10. [10]
    R. Mooney S. Bennett ”A domain independent explanation-based generalizer” In Proceedings AAAI 86 1986Google Scholar
  11. [11]
    P.S. Rosenbloom, J. Laird ”Mapping Explanation Based Generalization onto SOAR” in Proceedings AAAI 86, Philadelphia, PA. 1986Google Scholar
  12. [12]
    P.S. Rosenbloom, J. Laird, A. Newell ”Knowledge-level learning in SOAR.” in Proc. of Sixth National Conference on Artificial Intelligence Seattle. 1986Google Scholar
  13. [13]
    D. J. Scales ”Efficient Matching Algorithms for the SOAR / OPS5 Production System” Knowledge Systems Laboratory Report No KSL 86-47, 1986Google Scholar
  14. [14]
    M. Tambe, A. Newell ”Why Some chunks Are Expensive” Carnegie Mellon Computer Science Department, Report No 103. 1988Google Scholar
  15. [15]
    F. Zerr, J.G. Ganascia ”Comparaison du Chunking avec l'EBG implémenté sur SOAR” 5 ièmes Journées Françaises d'apprentissage, LANNION 1990Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1991

Authors and Affiliations

  • F. Zerr
    • 1
    • 2
  • J. G. Ganascia
    • 3
  1. 1.L.R.I.Université Parix XIOrsayFrance
  2. 2.THOMSON-CSF DSE/DT/SEP/STIBagneuxFrance
  3. 3.LAFORIA, URA 1095 du CNRSUniversité Pierre et Marie CURIEParisFrance

Personalised recommendations