A Structured Connectionist Approach to Inferencing and Retrieval

ROBIN, a structured (i.e., localist) connectionist model capable of massively-parallel high-level inferencing requiring variable bindings and rule application, and REMIND, a model based on ROBIN that explores the integration of language understanding and episodic memory retrieval in a single spreading-activation mechanism are described.

Abstract

Simple connectionist models have generally been unable to perform natural language understanding or memory retrieval beyond simple stereotypical situations that they have seen before. This is because they have had difficulties representing and applying general knowledge rules that specifically require variables, barring them from performing the high-level inferencing necessary for planning, reasoning, and natural language understanding. This chapter describes ROBIN, a structured (i.e., localist) connectionist model capable of massively-parallel high-level inferencing requiring variable bindings and rule application, and REMIND, a model based on ROBIN that explores the integration of language understanding and episodic memory retrieval in a single spreading-activation mechanism.