Processing

Please wait...

Settings

Settings

Goto Application

1. WO2022009005 - CONTINUAL LEARNING USING CROSS CONNECTIONS

Publication Number WO/2022/009005
Publication Date 13.01.2022
International Application No. PCT/IB2021/055392
International Filing Date 18.06.2021
IPC
G06N 3/04 2006.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
04Architecture, e.g. interconnection topology
CPC
G06N 20/00
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
20Machine learning
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 3/0445
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
0445Feedback networks, e.g. hopfield nets, associative networks
G06N 3/0454
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
0454using a combination of multiple neural nets
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
  • INTERNATIONAL BUSINESS MACHINES CORPORATION [US]/[US]
  • IBM UNITED KINGDOM LIMITED [GB]/[GB] (MG)
  • IBM (CHINA) INVESTMENT COMPANY LIMITED [CN]/[CN] (MG)
Inventors
  • TIAN, Yu
  • WANG, Xiaolong
  • XUE, Qinghan
  • ELIUK, Steven, Nicholas
  • GUO, Xin
Agents
  • SHAW, Anita
Priority Data
16/923,19608.07.2020US
Publication Language English (en)
Filing Language English (EN)
Designated States
Title
(EN) CONTINUAL LEARNING USING CROSS CONNECTIONS
(FR) APPRENTISSAGE CONTINUEL À L’AIDE DE CONNEXIONS TRANSVERSALES
Abstract
(EN) A method, a computer system, and a computer program product for using distinct paths with cross connections for distinct tasks to prevent catastrophic forgetting in class-incremental scenarios. The method may include: receiving one or more tasks sequentially; applying one or more shareable blocks to the one or more tasks; learning one or more distinct paths for the one or more tasks; adding one or more cross connections between the one or more tasks; adding an aggregation block to collect one or more outputs from the distinct paths of each of the one or more tasks; providing a prediction.
(FR) L'invention concerne un procédé, un système informatique, et un produit de programme informatique destinés à utiliser des chemins distincts avec des connexions transversales pour des tâches distinctes afin de prévenir l’oubli catastrophique dans des scenarios incrémentaux par classes. Le procédé peut comprendre les étapes consistant à: recevoir séquentiellement une ou plusieurs tâches; appliquer un ou plusieurs blocs partageables à la ou aux tâches; apprendre un ou plusieurs chemins distincts pour la ou les tâches; ajouter une ou plusieurs connexions transversales entre la ou les tâches; ajouter un bloc d’agrégation pour recueillir une ou plusieurs sorties provenant des chemins distincts de la ou de chacune des tâches; fournir une prédiction.
Related patent documents
Latest bibliographic data on file with the International Bureau