skip to main content
Visitante
Meu Espaço
Minha Conta
Sair
Identificação
This feature requires javascript
Tags
Revistas Eletrônicas (eJournals)
Livros Eletrônicos (eBooks)
Bases de Dados
Bibliotecas USP
Ajuda
Ajuda
Idioma:
Inglês
Espanhol
Português
This feature required javascript
This feature requires javascript
Primo Search
Busca Geral
Busca Geral
Acervo Físico
Acervo Físico
Produção Intelectual da USP
Produção USP
Search For:
Clear Search Box
Search in:
Busca Geral
Or hit Enter to replace search target
Or select another collection:
Search in:
Busca Geral
Busca Avançada
Busca por Índices
This feature requires javascript
This feature requires javascript
Structured Pruning of Deep Convolutional Neural Networks
Anwar, Sajid ; Hwang, Kyuyeon ; Sung, Wonyong
ACM journal on emerging technologies in computing systems, 2017-07, Vol.13 (3), p.1-18
[Periódico revisado por pares]
Texto completo disponível
Citações
Citado por
Exibir Online
Detalhes
Resenhas & Tags
Mais Opções
Nº de Citações
This feature requires javascript
Enviar para
Adicionar ao Meu Espaço
Remover do Meu Espaço
E-mail (máximo 30 registros por vez)
Imprimir
Link permanente
Referência
EasyBib
EndNote
RefWorks
del.icio.us
Exportar RIS
Exportar BibTeX
This feature requires javascript
Título:
Structured Pruning of Deep Convolutional Neural Networks
Autor:
Anwar, Sajid
;
Hwang, Kyuyeon
;
Sung, Wonyong
É parte de:
ACM journal on emerging technologies in computing systems, 2017-07, Vol.13 (3), p.1-18
Descrição:
Real-time application of deep learning algorithms is often hindered by high computational complexity and frequent memory accesses. Network pruning is a promising technique to solve this problem. However, pruning usually results in irregular network connections that not only demand extra representation efforts but also do not fit well on parallel computation. We introduce structured sparsity at various scales for convolutional neural networks: feature map-wise, kernel-wise, and intra-kernel strided sparsity. This structured sparsity is very advantageous for direct computational resource savings on embedded computers, in parallel computing environments, and in hardware-based systems. To decide the importance of network connections and paths, the proposed method uses a particle filtering approach. The importance weight of each particle is assigned by assessing the misclassification rate with a corresponding connectivity pattern. The pruned network is retrained to compensate for the losses due to pruning. While implementing convolutions as matrix products, we particularly show that intra-kernel strided sparsity with a simple constraint can significantly reduce the size of the kernel and feature map tensors. The proposed work shows that when pruning granularities are applied in combination, we can prune the CIFAR-10 network by more than 70% with less than a 1% loss in accuracy.
Idioma:
Inglês
This feature requires javascript
This feature requires javascript
Voltar para lista de resultados
This feature requires javascript
This feature requires javascript
Buscando em bases de dados remotas. Favor aguardar.
Buscando por
em
scope:(USP_VIDEOS),scope:("PRIMO"),scope:(USP_FISICO),scope:(USP_EREVISTAS),scope:(USP),scope:(USP_EBOOKS),scope:(USP_PRODUCAO),primo_central_multiple_fe
Mostrar o que foi encontrado até o momento
This feature requires javascript
This feature requires javascript