Loading...
Loading...

Go to the content (press return)

An out-of-the-box full-network embedding for convolutional neural networks

Author
Garcia-Gasulla, D.; Vilalta, A.; Parés, F.; Ayguade, E.; Labarta, J.; Cortés, U.; Suzumura, T.
Type of activity
Presentation of work at congresses
Name of edition
2018 IEEE International Conference on Big Knowledge
Date of publication
2018
Presentation's date
2018-11-17
Book of congress proceedings
2018 IEEE International Conference on Big Knowledge (ICBK 2018): Singapore 17-18 November 2018
First page
168
Last page
175
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
DOI
10.1109/ICBK.2018.00030
Project funding
Computación de Altas Prestaciones VII
Models de Programacio i Entorns d'eXecució PARal.lels
Repository
http://hdl.handle.net/2117/133555 Open in new window
URL
https://ieeexplore.ieee.org/document/8588789 Open in new window
Abstract
Features extracted through transfer learning can be used to exploit deep learning representations in contexts where there are very few training samples, where there are limited computational resources, or when the tuning of hyper-parameters needed for training deep neural networks is unfeasible. In this paper we propose a novel feature extraction embedding called full-network embedding. This embedding is based on two main points. First, the use of all layers in the network, integrating activatio...
Citation
Garcia-Gasulla, D. [et al.]. An out-of-the-box full-network embedding for convolutional neural networks. A: IEEE International Conference on Big Knowledge. "2018 IEEE International Conference on Big Knowledge (ICBK 2018): Singapore 17-18 November 2018". Institute of Electrical and Electronics Engineers (IEEE), 2018, p. 168-175.
Keywords
Embedding spaces, Feature extraction, Transfer learning
Group of research
CAP - High Performace Computing Group

Participants

  • Garcia Gasulla, Dario  (author and speaker )
  • Vilalta, Armand  (author and speaker )
  • Parés, Ferran  (author and speaker )
  • Ayguade Parra, Eduard  (author and speaker )
  • Labarta Mancho, Jesus Jose  (author and speaker )
  • Cortés, Ulises  (author and speaker )
  • Suzumura, Toyotaro  (author and speaker )