HETEROGENEOUS KNOWLEDGE REPRESENTATION - INTEGRATING CONNECTIONIST AND SYMBOLIC COMPUTATION

Heterogeneous knowledge representation allows combination of several k nowledge representation techniques. For instance, connectionist and sy mbolic systems are two different computational paradigms and knowledge representations. Unfortunately, the integration of different paradigm s and knowledge representations is not easy and very often is informal . In this paper, we propose a formal approach to integrate these two p aradigms where as a symbolic system we consider a (logic) rule-based s ystem. The integration is operated at language level between neural ne tworks and rule languages. The formal model that allows the integratio n is based on constraint logic programming and provides an integrated framework to represent and process heterogeneous knowledge. In order t o achieve this we define a new language that allows expression and mod elling in a natural and intuitive way the above issues together with t he operational semantics. (C) 1997 Elsevier Science B.V.