Processing

Please wait...

Settings

Settings

Goto Application

1. WO2020109891 - DECENTRALIZED DISTRIBUTED DEEP LEARNING

Publication Number WO/2020/109891
Publication Date 04.06.2020
International Application No. PCT/IB2019/059474
International Filing Date 05.11.2019
IPC
G06N 3/08 2006.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
CPC
G06F 17/16
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
17Digital computing or data processing equipment or methods, specially adapted for specific functions
10Complex mathematical operations
16Matrix or vector computation ; , e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
G06F 9/4881
GPHYSICS
06COMPUTING; CALCULATING; 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
4806Task transfer initiation or dispatching
4843by program, e.g. task dispatcher, supervisor, operating system
4881Scheduling strategies for dispatcher, e.g. round robin, multi-level priority queues
G06F 9/54
GPHYSICS
06COMPUTING; CALCULATING; 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
54Interprogram communication
G06K 9/6215
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6201Matching; Proximity measures
6215Proximity measures, i.e. similarity or distance measures
G06K 9/6257
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6217Design or setup of recognition systems and techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
6256Obtaining sets of training patterns; Bootstrap methods, e.g. bagging, boosting
6257characterised by the organisation or the structure of the process, e.g. boosting cascade
G06N 20/00
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
20Machine learning
Applicants
  • INTERNATIONAL BUSINESS MACHINES CORPORATION [US]/[US]
  • IBM UNITED KINGDOM LIMITED [GB]/[GB] (MG)
  • IBM (CHINA) INVESTMENT COMPANY LIMITED [CN]/[CN] (MG)
Inventors
  • ZHANG, Wei
  • ZHANG, Li
  • FINKLER, Ulrich
  • CHO, Minsik
  • KUNG, David
Agents
  • GASCOYNE, Belinda
Priority Data
16/206,27430.11.2018US
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) DECENTRALIZED DISTRIBUTED DEEP LEARNING
(FR) APPRENTISSAGE EN PROFONDEUR RÉPARTI ET DÉCENTRALISÉ
Abstract
(EN)
Various embodiments are provided for decentralized distributed deep learning by one or more processors in a computing system. Asynchronous distributed training of one or more machine learning models may be performed by generating a list of neighbour nodes for each node in a plurality of nodes and creating a first thread for continuous communication according to a weight management operation and a second thread for continuous computation of a gradient for each node. One or more variables are shared between the first thread and the second thread.
(FR)
Divers modes de réalisation concernent un apprentissage en profondeur réparti et décentralisé par un ou plusieurs processeurs dans un système informatique. Un apprentissage réparti asynchrone d'un ou de plusieurs modèles d'apprentissage automatique peut être réalisé en générant une liste de nœuds voisins pour chaque nœud d’une pluralité de nœuds et en créant un premier fil pour une communication continue selon une opération de gestion de poids, ainsi qu’un second fil pour le calcul continu d'un gradient pour chaque nœud. Une ou plusieurs variables sont partagées entre le premier fil et le second fil.
Also published as
Latest bibliographic data on file with the International Bureau