SCHEMAS, LOGICS, AND NEURAL ASSEMBLIES

To implement schemas and logics in connectionist models, some form of basic-level organization is needed. This paper proposes such an organi zation, which is termed a discrete neural assembly. Each discrete neur al assembly is in turn made up of discrete neurons (nodes), that is, a node that processes inputs based on a discrete mapping instead of a c ontinuous function. A group of discrete neurons (nodes) closely interc onnected form an assembly and carry out a basic functionality. Some su bstructures and superstructures of such assemblies are developed to en able complex symbolic schemas to be represented and processed in conne ctionist networks. The paper shows that logical inference can be perfo rmed precisely, when necessary, in these networks and with certain gen eralization, more flexible inference (fuzzy inference) can also be per formed. The development of various connectionist constructs demonstrat es the possibility of implementing symbolic schemas, in their full com plexity, in connectionist networks.