skip to main content
Idioma:
Primo Search
Clear Search Box
Search in: Busca Geral
Busca Avançada
Busca por Índices

Learning Architectures for Binary Networks

Vedaldi, Andrea ; Bischof, Horst ; Brox, Thomas ; Frahm, Jan-Michael

Computer Vision - ECCV 2020, 2020, Vol.12357, p.575-591 [Periódico revisado por pares]

Switzerland: Springer International Publishing AG

Sem texto completo

Citações Citado por
  • Título:
    Learning Architectures for Binary Networks
  • Autor: Vedaldi, Andrea ; Bischof, Horst ; Brox, Thomas ; Frahm, Jan-Michael
  • Assuntos: Architecture search ; Backbone architecture ; Binary networks
  • É parte de: Computer Vision - ECCV 2020, 2020, Vol.12357, p.575-591
  • Notas: D. Kim and K. P. Singh—Indicates equal contribution. This work is done while KPS is at GIST for internship.
    Electronic supplementary materialThe online version of this chapter (https://doi.org/10.1007/978-3-030-58610-2_34) contains supplementary material, which is available to authorized users.
  • Descrição: Backbone architectures of most binary networks are well-known floating point (FP) architectures such as the ResNet family. Questioning that the architectures designed for FP networks might not be the best for binary networks, we propose to search architectures for binary networks (BNAS) by defining a new search space for binary architectures and a novel search objective. Specifically, based on the cell based search method, we define the new search space of binary layer types, design a new cell template, and rediscover the utility of and propose to use the Zeroise layer instead of using it as a placeholder. The novel search objective diversifies early search to learn better performing binary architectures. We show that our method searches architectures with stable training curves despite the quantization error inherent in binary networks. Quantitative analyses demonstrate that our searched architectures outperform the architectures used in state-of-the-art binary networks and outperform or perform on par with state-of-the-art binary networks that employ various techniques other than architectural changes.
  • Títulos relacionados: Lecture Notes in Computer Science
  • Editor: Switzerland: Springer International Publishing AG
  • Idioma: Inglês
Voltar para lista de resultados

  • Realização: ABCD-USP USP   Logos de Redes Sociais: Freepik

Buscando em bases de dados remotas. Favor aguardar.