• xmlui.mirage2.page-structure.header.title
    • français
    • English
  • Help
  • Login
  • Language 
    • Français
    • English
View Item 
  •   BIRD Home
  • LAMSADE (UMR CNRS 7243)
  • LAMSADE : Publications
  • View Item
  •   BIRD Home
  • LAMSADE (UMR CNRS 7243)
  • LAMSADE : Publications
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Browse

BIRDResearch centres & CollectionsBy Issue DateAuthorsTitlesTypeThis CollectionBy Issue DateAuthorsTitlesType

My Account

LoginRegister

Statistics

Most Popular ItemsStatistics by CountryMost Popular Authors
Thumbnail

Training Compact Deep Learning Models for Video Classification Using Circulant Matrices

Araújo, Alexandre; Negrevergne, Benjamin; Chevaleyre, Yann; Atif, Jamal (2018), Training Compact Deep Learning Models for Video Classification Using Circulant Matrices, in Leal-Taixé, Laura; Roth, Stefan, Computer Vision – ECCV 2018 Workshops, Proceedings, Springer : Berlin Heidelberg, p. 271-286. 10.1007/978-3-030-11018-5

View/Open
2018_ECCV.pdf (298.8Kb)
Type
Communication / Conférence
Date
2018
Conference title
15th European Conference on Computer Vision (ECCV 2018)
Conference date
2018-09
Conference city
Munich
Conference country
Germany
Book title
Computer Vision – ECCV 2018 Workshops, Proceedings
Book author
Leal-Taixé, Laura; Roth, Stefan
Publisher
Springer
Published in
Berlin Heidelberg
ISBN
978-3-030-11017-8
Pages
271-286
Publication identifier
10.1007/978-3-030-11018-5
Metadata
Show full item record
Author(s)
Araújo, Alexandre
Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Negrevergne, Benjamin cc
Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Chevaleyre, Yann
Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Atif, Jamal
Laboratoire d'analyse et modélisation de systèmes pour l'aide à la décision [LAMSADE]
Abstract (EN)
In real world scenarios, model accuracy is hardly the only factor to consider. Large models consume more memory and are computationally more intensive, which make them difficult to train and to deploy, especially on mobile devices. In this paper, we build on recent results at the crossroads of Linear Algebra and Deep Learning which demonstrate how imposing a structure on large weight matrices can be used to reduce the size of the model. Building on these results, we propose very compact models for video classification based on state-of-the-art network architectures such as Deep Bag-of-Frames, NetVLAD and NetFisherVectors. We then conduct thorough experiments using the large YouTube-8M video classification dataset. As we will show, the circulant DBoF embedding achieves an excellent trade-off between size and accuracy.
Subjects / Keywords
Deep learning; Computer vision; Structured matrices; Circulant matrices

Related items

Showing items related by title and author.

  • Thumbnail
    On the Expressive Power of Deep Fully Circulant Neural Networks 
    Araújo, Alexandre; Negrevergne, Benjamin; Chevaleyre, Yann; Atif, Jamal (2019) Document de travail / Working paper
  • Thumbnail
    On Lipschitz Regularization of Convolutional Layers using Toeplitz Matrix Theory 
    Araújo, Alexandre; Negrevergne, Benjamin; Chevaleyre, Yann; Atif, Jamal (2021) Communication / Conférence
  • Thumbnail
    Building Compact and Robust Deep Neural Networks with Toeplitz Matrices 
    Araujo, Alexandre (2021-06-01) Thèse
  • Thumbnail
    Deep Learning for Metagenomic Data: using 2D Embeddings and Convolutional Neural Networks 
    Thanh Hai, Nguyen; Chevaleyre, Yann; Prifti, Edi; Sokolovska, Nataliya; Zucker, Jean-Daniel (2017) Communication / Conférence
  • Thumbnail
    Disease Classification in Metagenomics with 2D Embeddings and Deep Learning 
    Nguyen, Thanh Hai; Prifti, Edi; Chevaleyre, Yann; Sokolovska, Nataliya; Zucker, Jean-Daniel (2018) Communication / Conférence
Dauphine PSL Bibliothèque logo
Place du Maréchal de Lattre de Tassigny 75775 Paris Cedex 16
Phone: 01 44 05 40 94
Contact
Dauphine PSL logoEQUIS logoCreative Commons logo