Processing

Please wait...

PATENTSCOPE will be unavailable a few hours for maintenance reason on Tuesday 26.10.2021 at 12:00 PM CEST
Settings

Settings

Goto Application

1. WO2020139179 - A WIRELESS DEVICE, A NETWORK NODE AND METHODS THEREIN FOR TRAINING OF A MACHINE LEARNING MODEL

Publication Number WO/2020/139179
Publication Date 02.07.2020
International Application No. PCT/SE2018/051372
International Filing Date 28.12.2018
IPC
G06N 20/00 2019.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
20Machine learning
H03M 7/30 2006.1
HELECTRICITY
03BASIC ELECTRONIC CIRCUITRY
MCODING, DECODING OR CODE CONVERSION, IN GENERAL
7Conversion of a code where information is represented by a given sequence or number of digits to a code where the same information is represented by a different sequence or number of digits
30Compression; Expansion; Suppression of unnecessary data, e.g. redundancy reduction
CPC
G06N 20/00
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
20Machine learning
G06N 3/08
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
G06N 7/005
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
7Computer systems based on specific mathematical models
005Probabilistic networks
H03M 7/3059
HELECTRICITY
03BASIC ELECTRONIC CIRCUITRY
MCODING; DECODING; CODE CONVERSION IN GENERAL
7Conversion of a code where information is represented by a given sequence or number of digits to a code where the same ; , similar or subset of; information is represented by a different sequence or number of digits
30Compression
3059Digital compression and data reduction techniques where the original information is represented by a subset or similar information, e.g. lossy compression
H04L 67/04
HELECTRICITY
04ELECTRIC COMMUNICATION TECHNIQUE
LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
67Network-specific arrangements or communication protocols supporting networked applications
04adapted for terminals or networks with limited resources or for terminal portability, e.g. wireless application protocol [WAP]
H04L 69/04
HELECTRICITY
04ELECTRIC COMMUNICATION TECHNIQUE
LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
69Application independent communication protocol aspects or techniques in packet data networks
04Protocols for data compression
Applicants
  • TELEFONAKTIEBOLAGET LM ERICSSON (PUBL) [SE]/[SE]
Inventors
  • TULLBERG, Hugo
  • OTTERSTEN, Johan
Agents
  • AYOUB, Nabil
Priority Data
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) A WIRELESS DEVICE, A NETWORK NODE AND METHODS THEREIN FOR TRAINING OF A MACHINE LEARNING MODEL
(FR) DISPOSITIF SANS FIL, UN NŒUD DE RÉSEAU ET DES PROCÉDÉS ASSOCIÉS POUR ENTRAÎNER UN MODÈLE D’APPRENTISSAGE AUTOMATIQUE
Abstract
(EN)
A wireless device (120) and a method therein for assisting a network node (110) to perform training of a machine learning model. The wireless device collects a number of successive data samples. Further, the wireless device successively creates compressed data by associating each collected data sample to a cluster. The cluster has a cluster centroid, a cluster counter representative of a number of collected data samples determined to be normal and being associated with the cluster, and a number of outlier collected data samples associated with the cluster. Then, the wireless device updates the cluster centroid to correspond to a mean position of all normal data samples that are associated with the cluster, and increases the cluster counter by one for each normal data sample that is associated with the cluster. The wireless device transmits the compressed data to the network node.
(FR)
La présente invention concerne un dispositif sans fil (120) et un procédé associé pour aider un nœud de réseau (110) à réaliser un entraînement d’un modèle d’apprentissage automatique. Le dispositif sans fil collecte un certain nombre d’échantillons de données successifs. En outre, le dispositif sans fil crée successivement des données compressées par association de chaque échantillon de données collecté à un groupe. Le groupe a un centroïde, un compteur représentatif d’un certain nombre d’échantillons de données collectés, qui sont déterminés comme étant normaux et sont associés au groupe, et un certain nombre d’échantillons de données collectés anormaux associés au groupe. Ensuite, le dispositif sans fil met à jour le centroïde du groupe pour qu’il corresponde à une position moyenne de tous les échantillons de données normaux qui sont associés au groupe, et augmente le compteur du groupe d’une unité pour chaque échantillon de données normal qui est associé au groupe. Le dispositif sans fil transmet les données compressées au nœud de réseau.
Related patent documents
EP2018944874This application is not viewable in PATENTSCOPE because the national phase entry has not been published yet or the national entry is issued from a country that does not share data with WIPO or there is a formatting issue or an unavailability of the application.
Latest bibliographic data on file with the International Bureau