Processing

Please wait...

Settings

Settings

Goto Application

1. WO2021061962 - TRANSFER LEARNING FOR NEURAL NETWORKS

Publication Number WO/2021/061962
Publication Date 01.04.2021
International Application No. PCT/US2020/052464
International Filing Date 24.09.2020
IPC
G06N 3/04 2006.01
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
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
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 3/082
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
3Computer systems based on biological models
02using neural network models
08Learning methods
082modifying the architecture, e.g. adding or deleting nodes or connections, pruning
G06N 5/04
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
5Computer systems using knowledge-based models
04Inference methods or devices
Applicants
  • NVIDIA CORPORATION [US]/[US]
Inventors
  • AGHDASI, Farzin
  • PRAVEEN, Varun
  • RATNESH KUMAR, Fnu
  • SRIRAM, Partha
Agents
  • LOHR, Jason
Priority Data
17/029,72523.09.2020US
62/906,05425.09.2019US
Publication Language English (EN)
Filing Language English (EN)
Designated States
Title
(EN) TRANSFER LEARNING FOR NEURAL NETWORKS
(FR) APPRENTISSAGE DE TRANSFERT POUR DES RÉSEAUX À NEURONES
Abstract
(EN)
Transfer learning can be used to enable a user to obtain a machine learning model that is fully trained for an intended inferencing task without having to train the model from scratch. A pre-trained model can be obtained that is relevant for that inferencing task. Additional training data, as may correspond to at least one additional class of data, can be used to further train this model. This model can then be pruned and retrained in order to obtain a smaller model that retains high accuracy for the intended inferencing task.
(FR)
Selon la présente invention, un apprentissage de transfert peut être utilisé pour permettre à un utilisateur d’obtenir un modèle d’apprentissage automatique, qui est entièrement entraîné pour une tâche d’inférence prévue, sans devoir entraîner le modèle en partant de zéro. Il est possible d’obtenir un modèle pré-entraîné qui est pertinent pour cette tâche d’inférence. Des données d’entraînement supplémentaires, pouvant correspondre à au moins une classe supplémentaire de données, peuvent être utilisées pour un entraînement supplémentaire de ce modèle. Ce modèle peut ensuite être élagué et ré-entraîné afin d’obtenir un modèle plus petit qui conserve une grande précision pour la tâche d’inférence prévue.
Latest bibliographic data on file with the International Bureau