Processing

Please wait...

Settings

Settings

Goto Application

1. WO1996042067 - IMPROVED SELF-ORGANIZING CIRCUITS

Publication Number WO/1996/042067
Publication Date 27.12.1996
International Application No. PCT/US1995/007363
International Filing Date 08.06.1995
Chapter 2 Demand Filed 26.02.1996
IPC
G06K 9/66 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
64using simultaneous comparisons or correlations of the image signals with a plurality of references, e.g. resistor matrix
66references adjustable by an adaptive method, e.g. learning
CPC
G06K 9/66
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
64using simultaneous comparisons or correlations of the image signals with a plurality of references, e.g. resistor matrix
66references adjustable by an adaptive method, e.g. learning
Applicants
  • BUCKLEY, B., Shawn [US]/[US]
Inventors
  • BUCKLEY, B., Shawn
Agents
  • RITCHIE, William, B.
Priority Data
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) IMPROVED SELF-ORGANIZING CIRCUITS
(FR) SYSTEME AUTO-ORGANISATEUR AMELIORE
Abstract
(EN)
A self-organizing system providing improved performance is composed of node subcircuits (20, 21A, 21B, 21C, and 21N) in two- or three-dimensional arrays of nodes which behave like neurons in the brain. Improvements in the learning rules use the time-filtered output of nodes to define memory traces. Spatial summation and spatial difference functions then determine how node branches will compete to produce various memory trace topologies such as roots and junctions. By modifying node properties and branch competition of nodes at root junctions, roots interact to create Boolean logic roles.
(FR)
Ce système auto-organisateur, qui donne des résultats améliorés, se compose de sous-circuits de nodes (20, 21A, 21B, 21C et 21N) disposés dans des ensembles à deux ou trois dimensions de nodes qui se comportent comme les neurones du cerveau. Des améliorations des règles d'apprentissage proviennent de l'utilisation de données de sortie des nodes à filtrage temporel pour définir des trajets de mémoire. Des fonctions de sommations spatiales et de différences spatiales déterminent alors comment des branches de nodes entrent en concurrence pour produire diverses topologies de trajets de mémoires telles que des racines et des jonctions. En modifiant des propriétés nodales et la concurrence entre branches de nodes aux jonctions de racines, on fait interagir des racines pour qu'elles créent des rôles propres à la logique booléenne.
Also published as
Latest bibliographic data on file with the International Bureau