Investigation of Neural Network Dynamics

Abstract

The purpose of this document is to report the progress made in the first year of the grant entitled Investigation of Neural Network Dynamics. The proposed period of the work was September 1, 1987 to August 31, 1990, The proposed three year budget was $126,200 with a first year budget of $40,000. The grant was closed after a single year because the principal investigator moved from the Applied Physics Laboratory, California Institute of Technology. Nevertheless, many of the initial objectives were met in the single year that the grant was in force. The major result of this investigation is a systematic approach for exploiting the dynamics of a general class of neurodynamical systems for the purpose of neural computation. We have interpreted the back- propagation formalism as an adaptive algorithm for a general class of dynamical systems. The completely continuous formalism lends itself to implementation in analog VLSI. This can be accomplished without external synchronization.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 28, 1989
Accession Number
ADA216791

Entities

People

  • Fernando J. Pineda

Organizations

  • Johns Hopkins University

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Air Force
  • Algorithms
  • Artificial Intelligence Computing
  • California
  • Computational Complexity
  • Computations
  • Computer Science
  • Computers
  • Difference Equations
  • Differential Equations
  • Information Systems
  • Jet Propulsion
  • Neural Networks
  • Physics
  • Physics Laboratories
  • Recurrent Neural Networks
  • Students

Readers

  • Academic Conference Management
  • Control Systems Engineering.
  • Neural Network Machine Learning.

Technology Areas

  • AI & ML