THE MODELING OF RANDOM NEURAL NETWORKS ON AN ELECTRONIC DIGITAL COMPUTER,

Abstract

The properties and possible modes of functioning of a system consisting of set of interconnected quasi-neurons are examined. The system is modeled on a BESM-2M computer and consists of 1024 quasi neurons in one plane in the form of a 32 x 32 square. Each quasi-neuron has its own address and is excited if an only if the sum of excitations at its inputs exceeds the threshold value. The results indicate that the organization of couplings between quasi-neurons plays a substantial role in the functioning of nerve networks.

Document Details

Document Type
Technical Report
Publication Date
May 15, 1969
Accession Number
AD0697558

Entities

People

  • V. I. Lesis

Organizations

  • National Air and Space Intelligence Center

Tags

Communities of Interest

  • Air Platforms

DTIC Thesaurus Topics

  • Artificial Intelligence Computing
  • Artificial Intelligence Software
  • Computer Programs
  • Computers
  • Computing Devices
  • Couplings
  • Digital Computers
  • Excitation
  • Neural Networks

Readers

  • Distributed Systems and Data Platform Development
  • Neuroscience
  • Phased Array Antenna Design.

Technology Areas

  • AI & ML
  • AI & ML - Bayesian Inference
  • Microelectronics