Connectionist Models of Rule-Based Reasoning

We investigate connectionist models of rule-based reasoning, and show that while such models usually carry out reasoning in exactly the same way as symbolic systems, they have more to ooer in terms of commonsense reasoning. A connection-ist architecture for commonsense reasoning, CON-SYDERR, is proposed to account for common-sense reasoning patterns and to remedy the brit-tleness problem in traditional rule-based systems. A dual representational scheme is devised, utilizing both localist and distributed representations and exploring the synergy resulting from the interaction between the two. CONSYDERR is therefore capable of accounting for many diicult patterns in commonsense reasoning. This work shows that connectionist models of reasoning are not just \implementations" of their symbolic counterparts, but better computational models of commonsense reasoning.