BEELINE - A SITUATED, BOUNDED CONCEPTUAL KNOWLEDGE SYSTEM

How do recent theoretical arguments about situated cognition and bounded rationality affect the practical matter of designing and building intelligent knowledge based systems? A particular view of situated cognition is adopted, in which conceptual knowledge structures are grounded by being linked to objects, events and behaviours in nested environments outside the system through channels that are analogous to biological sensors and actuators. These grounded conceptual structures can be created and modified through a variety of arbitrarily bounded conceptual processes, including natural language text parsing. Fruitful interactions between conceptual structures from these different sources may then be explored. BEELINE is a software platform designed for experimenting with these ideas. It immerses a language-using, goal-seeking cognitive agent in a simulated physical world. It can use real English instructions to achieve a navigational goal. This predicament offers fresh solutions to old knowledge engineering problems, and presents some new challenges for knowledge based systems of the future.

[1]  Ronald J. Brachman,et al.  An Overview of the KL-ONE Knowledge Representation System , 1985, Cogn. Sci..

[2]  John F. Sowa,et al.  Using a lexicon of canonical graphs in a semantic interpreter , 1989 .

[3]  Kenneth M. Ford,et al.  On Babies and Bathwater: A Cautionary Tale , 1994, AI Mag..

[4]  J. Ross Quinlan,et al.  C4.5: Programs for Machine Learning , 1992 .

[5]  Douglas B. Lenat,et al.  Why AM and EURISKO Appear to Work , 1984, Artif. Intell..

[6]  Joachim Hoffmann,et al.  Continuous information processing in conceptual identifications , 1985, Psychological research.

[7]  Allen Newell,et al.  Human Problem Solving. , 1973 .

[8]  Stuart J. Russell.  An architecture for bounded rationality , 1991, SGAR.

[9]  John F. Sowa,et al.  Implementing a Semantic Interpreter Using Conceptual Graphs , 1986, IBM J. Res. Dev..

[10]  Pattie Maes,et al.  Agents that reduce work and information overload , 1994, CACM.

[11]  Luc Steels,et al.  Emergent Frame Recognition and Its Use in Artificial Creatures , 1991, IJCAI.

[12]  Stuart J. Russell,et al.  Planning Using Multiple Execution Architectures , 1993, IJCAI.

[13]  Rodney A. Brooks,et al.  A robot that walks; emergent behaviors from a carefully evolved network , 1989, Proceedings, 1989 International Conference on Robotics and Automation.

[14]  Herbert A. Simon,et al.  Situated Action: A Symbolic Interpretation , 1993, Cogn. Sci..

[15]  John F. Sowa,et al.  Conceptual graphs summary , 1992 .

[16]  Noam Chomsky.  Review of B.F. Skinner, Verbal Behavior , 1959 .

[17]  Richard Scaglion.  Cognition in practice: Mind, mathematics and culture in everyday life: Jean Lave (Cambridge University Press, Cambridge, 1988) pp. xv + 214, $14.95 (paper), $49.50 (cloth) , 1990 .

[18]  R. A. Brooks,et al.  Intelligence without Representation , 1991, Artif. Intell..

[19]  R. Peter Bonasso,et al.  Integrating Reaction Plans and Layered Competences Through Synchronous Control , 1991, IJCAI.

[20]  G. A. Miller.  THE PSYCHOLOGICAL REVIEW THE MAGICAL NUMBER SEVEN, PLUS OR MINUS TWO: SOME LIMITS ON OUR CAPACITY FOR PROCESSING INFORMATION 1 , 1956 .

[21]  Joseph E. Grimes,et al.  The Thread of Discourse , 1984 .