Accession Number : ADA336093

Title :   Biomorphic Networks for ATR and Higher-Level Processing

Descriptive Note : Quarterly rept. 1 Oct 97-1 Jan 98

Corporate Author : PENNSYLVANIA UNIV PHILADELPHIA

Personal Author(s) : Farbat, Nabil H.

PDF Url : ADA336093

Report Date : 10 JAN 1998

Pagination or Media Count : 10

Abstract : We introduce the concept of parametrically and nonlinearly coupled network of bifurcation processing elements that can be driven by static or dynamic input patterns. The network is biologically inspired, computes with all three types of attractors, and offers a unique tool for the modeling and study of cortical networks and higher level brain function.

Descriptors :   *NEURAL NETS, *BIFURCATION(BIOLOGY), NERVE CELLS, PATTERN RECOGNITION, CEREBRAL CORTEX, MAGNETIC RESONANCE IMAGING.

Subject Categories : Anatomy and Physiology
      Cybernetics

Distribution Statement : APPROVED FOR PUBLIC RELEASE