Advances in SHRUTI - a neurally motivated model of relational knowledge representation and rapid inference using temporal synchrony

摘要

SHRUTI is a connectionist model that demonstrates how a neurally plausible network can encode a large body of semantic and episodic facts, systematic rules, and knowledge about entities and types, and yet perform a wide range of explanatory and predictive inferences within a few hundred milliseconds. Over the past few years, this model has undergone enhancements. These enhancements enable SHRUTI to (1) deal with negation and inconsistent beliefs, (2) encode evidential rules and facts, (3) perform inferences requiring the dynamic instantation of entities, and (4) seek coherent explanations of observations.