Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020193329 - MACHINE LEARNING

Publication Number WO/2020/193329
Publication Date 01.10.2020
International Application No. PCT/EP2020/057529
International Filing Date 18.03.2020
IPC
G06N 5/00 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
5Computer systems using knowledge-based models
G06N 7/02 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
7Computer systems based on specific mathematical models
02using fuzzy logic
G06N 3/12 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
12using genetic models
CPC
G06N 3/126
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
12using genetic models
126Genetic algorithms, i.e. information processing using digital simulations of the genetic system
G06N 7/023
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
7Computer systems based on specific mathematical models
02using fuzzy logic
023Learning or tuning the parameters of a fuzzy system
Applicants
  • BRITISH TELECOMMUNICATIONS PUBLIC LIMITED COMPANY [GB]/[GB]
Inventors
  • OWUSU, Gilbert
  • HAGRAS, Hani
  • CHIMATAPU, Ravikiran
  • STARKEY, Andrew
Agents
  • BRITISH TELECOMMUNICATIONS PUBLIC LIMITED COMPANY, INTELLECTUAL PROPERTY DEPARTMENT
Priority Data
19164777.523.03.2019EP
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) MACHINE LEARNING
(FR) APPRENTISSAGE MACHINE
Abstract
(EN)
A computer implemented method for machine learning comprising: training an autoencoder having a set of input units, a set of output units and at least one set of hidden units, wherein connections between each of the sets of units are provided by way of interval type-2 fuzzy logic systems each including one or more rules, and the fuzzy logic systems are trained using an optimisation algorithm; and generating a representation of rules in each of the interval type-2 fuzzy logic systems triggered beyond a threshold by input data provided to the input units so as to indicate the rules involved in generating an output at the output units in response to the data provided to the input units.
(FR)
L'invention concerne un procédé mis en œuvre par ordinateur pour un apprentissage machine, qui comprend : l'apprentissage d'un autocodeur ayant un ensemble d'unités d'entrée, un ensemble d'unités de sortie et au moins un ensemble d'unités cachées, des connexions entre chacun des ensembles d'unités étant fournies au moyen de systèmes à logique floue de type 2 d'intervalle comprenant chacun une ou plusieurs règles, et les systèmes à logique floue étant entraînés à l'aide d'un algorithme d'optimisation; et la génération d'une représentation de règles dans chacun des systèmes à logique floue de type 2 d'intervalle déclenché au-delà d'un seuil par des données d'entrée fournies aux unités d'entrée de façon à indiquer les règles impliquées dans la génération d'une sortie au niveau des unités de sortie en réponse aux données fournies aux unités d'entrée.
Latest bibliographic data on file with the International Bureau