Ir al contenido (pulsa Retorno)

Universitat Politècnica de Catalunya

    • Català
    • Castellano
    • English
    • LoginRegisterLog in (no UPC users)
  • mailContact Us
  • world English 
    • Català
    • Castellano
    • English
  • userLogin   
      LoginRegisterLog in (no UPC users)

UPCommons. Global access to UPC knowledge

58.848 UPC E-Prints
You are here:
View Item 
  •   DSpace Home
  • E-prints
  • Grups de recerca
  • KEMLG - Grup d'Enginyeria del Coneixement i Aprenentatge Automàtic
  • Capítols de llibre
  • View Item
  •   DSpace Home
  • E-prints
  • Grups de recerca
  • KEMLG - Grup d'Enginyeria del Coneixement i Aprenentatge Automàtic
  • Capítols de llibre
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

On the representativeness of convolutional neural networks layers

Thumbnail
View/Open
On the Representativeness of.pdf (4,291Mb)
Share:
 
 
10.3233/978-1-61499-696-5-29
 
  View Usage Statistics
Cita com:
hdl:2117/96526

Show full item record
García Gasulla, Darío
Moreno, Jonatan
Ramos-Pollan, Raúl
Casadiegos Barrios, Romel
Béjar Alonso, JavierMés informacióMés informacióMés informació
Cortés García, Claudio UlisesMés informacióMés informacióMés informació
Ayguadé Parra, EduardMés informacióMés informacióMés informació
Labarta Mancho, Jesús JoséMés informacióMés informacióMés informació
Suzumura, Toyotaro
Document typePart of book or chapter of book
Defense date2016
PublisherIOS PRESS EBOOKS
Rights accessOpen Access
All rights reserved. This work is protected by the corresponding intellectual and industrial property rights. Without prejudice to any existing legal exemptions, reproduction, distribution, public communication or transformation of this work are prohibited without permission of the copyright holder
ProjectCOMPUTACION DE ALTAS PRESTACIONES VII (MINECO-TIN2015-65316-P)
Abstract
Convolutional Neural Networks (CNN) are the most popular of deep network models due to their applicability and success in image processing. Although plenty of effort has been made in designing and training better discriminative CNNs, little is yet known about the internal features these models learn. Questions like, what specific knowledge is coded within CNN layers, and how can it be used for other purposes besides discrimination, remain to be answered. To advance in the resolution of these questions, in this work we extract features from CNN layers, building vector representations from CNN activations. The resultant vector embedding is used to represent first images and then known image classes. On those representations we perform an unsupervised clustering process, with the goal of studying the hidden semantics captured in the embedding space. Several abstract entities untaught to the network emerge in this process, effectively defining a taxonomy of knowledge as perceived by the CNN. We evaluate and interpret these sets using WordNet, while studying the different behaviours exhibited by the layers of a CNN model according to their depth. Our results indicate that, while top (i.e., deeper) layers provide the most representative space, low layers also define descriptive dimensions.
CitationGarcía, D., Moreno, J., Ramos-Pollan, R., Barrios, R., Béjar, J., Cortés, C., Ayguadé, E., Labarta, J., Suzumura, T. On the representativeness of convolutional neural networks layers. A: "Artificial Intelligence Research and Development: proceedings of the 19th International Conference of the Catalan Association for Artificial Intelligence: Barcelona, Catalonia, Spain, October 19–21, 2016". IOS PRESS EBOOKS, 2016, p. 29-38. 
URIhttp://hdl.handle.net/2117/96526
DOI10.3233/978-1-61499-696-5-29
ISBN978-1-61499-695-8
Publisher versionhttp://ebooks.iospress.nl/volumearticle/45307
Collections
  • KEMLG - Grup d'Enginyeria del Coneixement i Aprenentatge Automàtic - Capítols de llibre [15]
  • CAP - Grup de Computació d'Altes Prestacions - Capítols de llibre [29]
  • Computer Sciences - Capítols de llibre [21]
  • Departament de Ciències de la Computació - Capítols de llibre [83]
  • Departament d'Arquitectura de Computadors - Capítols de llibre [75]
Share:
 
  View Usage Statistics

Show full item record

FilesDescriptionSizeFormatView
On the Representativeness of.pdf4,291MbPDFView/Open

Browse

This CollectionBy Issue DateAuthorsOther contributionsTitlesSubjectsThis repositoryCommunities & CollectionsBy Issue DateAuthorsOther contributionsTitlesSubjects

© UPC Obrir en finestra nova . Servei de Biblioteques, Publicacions i Arxius

info.biblioteques@upc.edu

  • About This Repository
  • Contact Us
  • Send Feedback
  • Inici de la pàgina