Synopsis A full rationale is presented for designing systems with enhanced cognitive properties using RAM-based or "weightless" methods. The paper concentrates on a structure with associative properties called the General Neural Unit (GNU) and shows that weightless analysis leads to an understanding of storage capacity and retrievability of stored training patterns in this unit. A discussion is included on the way that many such units might be used by a system designer in the making of cognitively competent machines.
Key words:
Neural modelling, RAM nodes, M-PLN's, reinforcement training, pattern retrieval, cognitive modelling.