Accession Number : ADA292785

Title :   Mathematical Theory of Neural Networks.

Descriptive Note : Final rept. 1 Aug 91-31 Jul 94,

Corporate Author : RUTGERS - THE STATE UNIV NEW BRUNSWICK NJ DEPT OF MATHEMATICS

Personal Author(s) : Sontag, Eduardo D. ; Sussmann, Hector J.

PDF Url : ADA292785

Report Date : 31 AUG 1994

Pagination or Media Count : 32

Abstract : This report provides a summary of the grant work by the principal investigators in the area of neural networks. The topics covered deal with (1) analysis of networks from the viewpoint of analog computational devices, exploring limitations imposed by resource constraints; (2) questions of parameter identification of recurrent nets; (3) use of feedforward nets for function approximation and interpolation problems; (4) systems theory (observability and other properties) for nets; and (5) the use of neural networks for the control of nonlinear systems.

Descriptors :   *NEURAL NETS, CONTROL SYSTEMS, OPTIMIZATION, COMPUTATIONS, PARAMETERS, THEORY, NUMERICAL ANALYSIS, IDENTIFICATION, NONLINEAR SYSTEMS, INTERPOLATION, MATHEMATICS, ANALOG SYSTEMS.

Subject Categories : Theoretical Mathematics
      Cybernetics

Distribution Statement : APPROVED FOR PUBLIC RELEASE