oru.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Learning feature representations with a cost-relevant sparse autoencoder
Örebro University, School of Science and Technology. (AASS)ORCID iD: 0000-0002-0579-7181
Örebro University, School of Science and Technology. (AASS)ORCID iD: 0000-0002-3122-693X
2015 (English)In: International Journal of Neural Systems, ISSN 0129-0657, E-ISSN 1793-6462, Vol. 25, no 1, p. 1450034-Article in journal (Refereed) Published
Abstract [en]

There is an increasing interest in the machine learning community to automatically learn feature representations directly from the (unlabeled) data instead of using hand-designed features. The autoencoder is one method that can be used for this purpose. However, for data sets with a high degree of noise, a large amount of the representational capacity in the autoencoder is used to minimize the reconstruction error for these noisy inputs. This paper proposes a method that improves the feature learning process by focusing on the task relevant information in the data. This selective attention is achieved by weighting the reconstruction error and reducing the influence of noisy inputs during the learning process. The proposed model is trained on a number of publicly available image data sets and the test error rate is compared to a standard sparse autoencoder and other methods, such as the denoising autoencoder and contractive autoencoder.

Place, publisher, year, edition, pages
2015. Vol. 25, no 1, p. 1450034-
Keywords [en]
Sparse autoencoder; unsupervised feature learning; weighted cost function
National Category
Other Engineering and Technologies Computer Engineering
Research subject
Computer Science
Identifiers
URN: urn:nbn:se:oru:diva-40063DOI: 10.1142/S0129065714500348ISI: 000347965500005PubMedID: 25515941OAI: oai:DiVA.org:oru-40063DiVA, id: diva2:774922
Available from: 2014-12-29 Created: 2014-12-29 Last updated: 2018-06-26Bibliographically approved
In thesis
1. Modeling time-series with deep networks
Open this publication in new window or tab >>Modeling time-series with deep networks
2014 (English)Doctoral thesis, comprehensive summary (Other academic)
Place, publisher, year, edition, pages
Örebro: Örebro university, 2014. p. 56
Series
Örebro Studies in Technology, ISSN 1650-8580 ; 63
Keywords
multivariate time-series, deep learning, representation learning, unsupervised
National Category
Computer and Information Sciences
Research subject
Information technology
Identifiers
urn:nbn:se:oru:diva-39415 (URN)978-91-7529-054-6 (ISBN)
Public defence
2015-02-02, Hörsalen, Musikhögskolan, Örebro universitet, Fakultetsgatan 1, Örebro, 13:15 (English)
Opponent
Supervisors
Available from: 2014-12-08 Created: 2014-12-08 Last updated: 2018-04-05Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full textPubMed

Authority records BETA

Längkvist, MartinLoutfi, Amy

Search in DiVA

By author/editor
Längkvist, MartinLoutfi, Amy
By organisation
School of Science and Technology
In the same journal
International Journal of Neural Systems
Other Engineering and TechnologiesComputer Engineering

Search outside of DiVA

GoogleGoogle Scholar

doi
pubmed
urn-nbn

Altmetric score

doi
pubmed
urn-nbn
Total: 415 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf