Advances in SHRUTI—A Neurally Motivated Model of RelationalKnowledge Representation and Rapid Inference Using Temporal Synchrony

  • Authors:
  • Lokendra Shastri

  • Affiliations:
  • International Computer Science Institute, 1947 Center Street, Suite 600, Berkeley, CA 94704. www.icsi.berkeley.edu/∼shastri

  • Venue:
  • Applied Intelligence
  • Year:
  • 1999

Quantified Score

Hi-index 0.00

Visualization

Abstract

We are capable of drawing a variety of inferences effortlessly, spontaneously, and with remarkable efficiency—as though these inferences are a reflex response of our cognitive apparatus. This remarkable human ability poses achallenge for cognitive science and computational neuroscience: How can a network of slow neuron-like elements represent a large body of systematic knowledge and perform a wide range of inferences with such speed? The connectionist model SHRUTI attempts to address thischallenge by demonstrating how a neurally plausible network can encodea large body of semantic and episodic facts, systematic rules,and knowledge about entities and types, and yet perform a wide range ofexplanatory and predictive inferences within a few hundred milliseconds.Relational structures (frames, schemas) are represented in SHRUTIby clusters of cells, and inference in SHRUTI correspondsto a transient propagation of rhythmic activity over such cell-clusterswherein dynamic bindings are represented by the synchronous firingof appropriate cells. SHRUTI encodes mappings acrossrelational structures using high-efficacy links that enable the propagationof rhythmic activity, and it encodes items in long-term memory ascoincidence and coincidence-error detector circuits that become activein response to the occurrence (or non-occurrence) of appropriatecoincidences in the on going flux of rhythmic activity. Finally, “understanding” in SHRUTI corresponds toreverberant and coherent activity along closed loops of neural circuitry.Over the past several years, SHRUTI has undergone severalenhancements that have augmented its expressiveness and inferential power.This paper describes some of these extensions that enable SHRUTIto (i) deal with negation and inconsistent beliefs, (ii) encode evidentialrules and facts, (iii) perform inferences requiring the dynamic instantiationof entities, and (iv) seek coherent explanations of observations.