Compression progressive et renforcement ...
Type de document :
Compte-rendu et recension critique d'ouvrage
DOI :
Titre :
Compression progressive et renforcement du poids pour les réseaux de neurones a impulsions
Auteur(s) :
Elbez, Hammouda [Auteur]
Benhaoua, Kamel [Auteur]
Devienne, Philippe [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Boulet, Pierre [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Benhaoua, Kamel [Auteur]
Devienne, Philippe [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Boulet, Pierre [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Titre de la revue :
Concurrency and Computation: Practice and Experience
Éditeur :
Wiley
Date de publication :
2022-03-02
ISSN :
1532-0626
Discipline(s) HAL :
Informatique [cs]/Intelligence artificielle [cs.AI]
Informatique [cs]/Apprentissage [cs.LG]
Informatique [cs]/Apprentissage [cs.LG]
Résumé en anglais : [en]
Neuromorphic architectures are one of the most promising architectures to significantly reduce the energy consumption of tomorrow’s computers. These architectures are inspired by the behaviour of the brain at a fairly ...
Lire la suite >Neuromorphic architectures are one of the most promising architectures to significantly reduce the energy consumption of tomorrow’s computers. These architectures are inspired by the behaviour of the brain at a fairly precise level and consist of artificial Spiking Neural Networks (SNNs). To optimise the implementation of these architectures, we propose in this paper a novel progressive network compression and reinforcement technique based on two functions, progressive pruning and dynamic synaptic weight reinforcement used after each training batch. The proposed approach delivers a highly compressed network (up to 75 % of compression rate) while preserving the network performance when tested with MNIST.Lire moins >
Lire la suite >Neuromorphic architectures are one of the most promising architectures to significantly reduce the energy consumption of tomorrow’s computers. These architectures are inspired by the behaviour of the brain at a fairly precise level and consist of artificial Spiking Neural Networks (SNNs). To optimise the implementation of these architectures, we propose in this paper a novel progressive network compression and reinforcement technique based on two functions, progressive pruning and dynamic synaptic weight reinforcement used after each training batch. The proposed approach delivers a highly compressed network (up to 75 % of compression rate) while preserving the network performance when tested with MNIST.Lire moins >
Langue :
Anglais
Vulgarisation :
Non
Collections :
Source :
Fichiers
- https://hal.archives-ouvertes.fr/hal-02737057/document
- Accès libre
- Accéder au document
- https://hal.archives-ouvertes.fr/hal-02737057/document
- Accès libre
- Accéder au document
- https://hal.archives-ouvertes.fr/hal-02737057/document
- Accès libre
- Accéder au document
- document
- Accès libre
- Accéder au document
- Final_backup.pdf
- Accès libre
- Accéder au document