[go: up one dir, main page]

CA2245571A1 - Procede de traitement de flux de donnees dans un reseau neuronal, et reseau neuronal - Google Patents

Procede de traitement de flux de donnees dans un reseau neuronal, et reseau neuronal Download PDF

Info

Publication number
CA2245571A1
CA2245571A1 CA 2245571 CA2245571A CA2245571A1 CA 2245571 A1 CA2245571 A1 CA 2245571A1 CA 2245571 CA2245571 CA 2245571 CA 2245571 A CA2245571 A CA 2245571A CA 2245571 A1 CA2245571 A1 CA 2245571A1
Authority
CA
Canada
Prior art keywords
input
output
neurons
neural network
parameter representation
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Abandoned
Application number
CA 2245571
Other languages
English (en)
Inventor
Rodney Michael John Cotterill
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority claimed from PCT/DK1997/000043 external-priority patent/WO1997030400A1/fr
Publication of CA2245571A1 publication Critical patent/CA2245571A1/fr
Abandoned legal-status Critical Current

Links

Landscapes

  • Image Analysis (AREA)

Abstract

Ce réseau neuronal possède un capteur extérieur (2) destiné à recevoir une représentation paramétrique de l'environnement. Cette représentation est pondérée et envoyée à l'entrée d'un noeud maître (1) dont la sortie transfère ladite représentation, sous une forme de nouveau pondérée, à l'entrée d'un effecteur extérieur (5). La sortie de cet effecteur est renvoyée à l'entrée du capteur, après influence sur l'environnement. En outre, cette représentation est renvoyée directement à partir du capteur du noeud maître, une somme des représentations paramétriques renvoyées s'effectuant ensuite. L'apprentissage par le réseau neuronal est rapide et très complexe, le réseau pouvant, après cet apprentissage, exécuter des tâches sans avoir à utiliser des signaux de retour influencés par l'environnement. Ainsi, l'invention permet d'obtenir un réseau neuronal présentant des analogies avec la conscience humaine.
CA 2245571 1996-02-02 1997-02-03 Procede de traitement de flux de donnees dans un reseau neuronal, et reseau neuronal Abandoned CA2245571A1 (fr)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
DK11196 1996-02-02
DK0111/96 1996-02-02
PCT/DK1997/000043 WO1997030400A1 (fr) 1996-02-02 1997-02-03 Procede de traitement de flux de donnees dans un reseau neuronal, et reseau neuronal

Publications (1)

Publication Number Publication Date
CA2245571A1 true CA2245571A1 (fr) 1997-08-21

Family

ID=29421777

Family Applications (1)

Application Number Title Priority Date Filing Date
CA 2245571 Abandoned CA2245571A1 (fr) 1996-02-02 1997-02-03 Procede de traitement de flux de donnees dans un reseau neuronal, et reseau neuronal

Country Status (1)

Country Link
CA (1) CA2245571A1 (fr)

Similar Documents

Publication Publication Date Title
Wang et al. The design of memristive circuit for affective multi-associative learning
Yamashita et al. Spontaneous prediction error generation in schizophrenia
Mpitsos et al. Convergence and divergence in neural networks: Processing of chaos and biological analogy
Magoulas et al. Neuro-fuzzy Synergism for Planning the Content in a Web-based Course
Grossberg Beta oscillations and hippocampal place cell learning during exploration of novel environments
JPH06506785A (ja) 視界の背景から動いている図を識別する装置
Fujii et al. Itinerant dynamics of class I* neurons coupled by gap junctions
Bosse et al. A cognitive and neural model for adaptive emotion reading by mirroring preparation states and Hebbian learning
Koprinkova-Hristova et al. Spike timing neural model of motion perception and decision making
EP0877984B1 (fr) Procede de traitement de flux de donnees dans un reseau neuronal, et reseau neuronal
Sarma et al. Intelligent tutoring systems using reinforcement learning to teach autistic students
Zins et al. Reproducing fear conditioning of rats with unmanned ground vehicles and neuromorphic systems
CA2245571A1 (fr) Procede de traitement de flux de donnees dans un reseau neuronal, et reseau neuronal
Zins et al. Implementation of Associative Memory Learning in Mobile Robots Using Neuromorphic Computing
Koprinkova-Hristova et al. Investigation of feedback connections effect of a spike timing neural network model of early visual system
Mengel et al. On the use of neural networks in intelligent tutoring systems
Postma et al. Robust Feedforward Processing in Syn re Chains
Bálya et al. A realistic mammalian retinal model implemented on complex cell CNN universal machine
Naveenraj et al. Spiking Neural Network-Based Digit Pattern Recognition in FPGA
Hutchison The central role for behavior analysis in modern robotics, and vice versa
Kinouchi et al. A Conscious AI System Based on Recurrent Neural Networks Applying Dynamic Information Equilibrium.
Weng et al. Online learning for attention, recognition, and tracking by a single developmental framework
Li et al. Training a kind of hybrid universal learning networks with classification problems
Lagriffoul et al. Perceiving and acting out of the box
Warner et al. Intervention informatics-healing with information

Legal Events

Date Code Title Description
FZDE Dead