Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020111647 - MULTI-TASK BASED LIFELONG LEARNING

Publication Number WO/2020/111647
Publication Date 04.06.2020
International Application No. PCT/KR2019/015949
International Filing Date 20.11.2019
IPC
G06N 20/00 2019.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
20Machine learning
G06F 9/48 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
FELECTRIC DIGITAL DATA PROCESSING
9Arrangements for program control, e.g. control units
06using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
46Multiprogramming arrangements
48Program initiating; Program switching, e.g. by interrupt
CPC
G06N 3/04
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architectures, e.g. interconnection topology
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
Applicants
  • SAMSUNG ELECTRONICS CO., LTD. [KR]/[KR]
Inventors
  • ZHANG, Jie
  • ZHANG, Junting
  • GHOSH, Shalini
  • LI, Dawei
  • ZHU, Jingwen
Agents
  • Y.P.LEE,MOCK & PARTNERS
Priority Data
16/379,70409.04.2019US
62/774,04330.11.2018US
62/792,05214.01.2019US
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) MULTI-TASK BASED LIFELONG LEARNING
(FR) APPRENTISSAGE CONTINU BASÉ SUR DES TÂCHES MULTIPLES
Abstract
(EN)
Methods, devices, and computer-readable media for multi-task based lifelong learning. A method for lifelong learning includes identifying a new task for a machine learning model to perform. The machine learning model trained to perform an existing task. The method includes adaptively training a network architecture of the machine learning model to generate an adapted machine learning model based on incorporating inherent correlations between the new task and the existing task. The method further includes using the adapted machine learning model to perform both the existing task and the new task.
(FR)
L'invention concerne des procédés, des dispositifs et des supports lisibles par ordinateur pour un apprentissage continu basé sur des tâches multiples. L'invention concerne également un procédé d'apprentissage continu qui consiste à identifier une nouvelle tâche pour un modèle d'apprentissage automatique à effectuer. Le modèle d'apprentissage machine est entraîné pour effectuer une tâche existante. Le procédé comprend l'apprentissage adaptatif d'une architecture de réseau du modèle d'apprentissage machine pour générer un modèle d'apprentissage machine adapté sur la base de l'incorporation de corrélations inhérentes entre la nouvelle tâche et la tâche existante. Le procédé comprend en outre l'utilisation du modèle d'apprentissage automatique adapté pour effectuer à la fois la tâche existante et la nouvelle tâche.
Also published as
Latest bibliographic data on file with the International Bureau