• DocumentCode
    3376484
  • Title

    A hybrid/symbolic connectionist production system

  • Author

    Sanou, Katsuaki ; Romaniuk, Steve G. ; Hall, Lawrence O.

  • Author_Institution
    Dept. of Comput. Sci. & Eng., Univ. of South Florida, Tampa, FL, USA
  • fYear
    1992
  • fDate
    10-13 Nov 1992
  • Firstpage
    44
  • Lastpage
    53
  • Abstract
    The task of implementing a simple rule-based production system in terms of a hybrid symbolic/connectionist architecture is discussed. The aim is to achieve parallel execution of rules. The architecture uses a local representation and builds upon prior work, which resulted in the symbolic/connectionist expert system development tool SC-net. The Hybrid Symbolic/Connectionist Production System (HSC-PS) reported supports two types of working memory elements, the attribute/value pair and the object/attribute/value triplet. HSC-PS can provide variable binding in an individual condition and across conditions and variable value instantiation from the left-hand side to the right-hand side in a rule. The network is constructed from production rules such that only one value is bound to a variable at any one time. An implementation of the farmer´s dilemma planning problem is given to illustrate the potential of this approach
  • Keywords
    inference mechanisms; neural nets; SC-net; attribute/value pair; expert system development tool; farmer´s dilemma planning problem; hybrid/symbolic connectionist production system; left-hand side; memory elements; object/attribute/value triplet; parallel execution; right-hand side; rule-based production system; variable value instantiation; Computer aided manufacturing; Computer architecture; Computer science; Hardware; Hybrid intelligent systems; Parallel processing; Production planning; Production systems; Robustness; Steel;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Tools with Artificial Intelligence, 1992. TAI '92, Proceedings., Fourth International Conference on
  • Conference_Location
    Arlington, VA
  • Print_ISBN
    0-8186-2905-3
  • Type

    conf

  • DOI
    10.1109/TAI.1992.246374
  • Filename
    246374