• English
  • Deutsch
  • Log In
    or
  • Research Outputs
  • Projects
  • Researchers
  • Institutes
  • Statistics
Repository logo
Fraunhofer-Gesellschaft
  1. Home
  2. Fraunhofer-Gesellschaft
  3. Konferenzschrift
  4. Utilizing Representation Learning for Robust Text Classification Under Datasetshift
 
  • Details
  • Full
Options
2021
Conference Paper
Titel

Utilizing Representation Learning for Robust Text Classification Under Datasetshift

Abstract
Within One-vs-Rest (OVR) classification, a classifier differentiates a single class of interest (COI) from the rest, i.e. any other class. By extending the scope of the rest class to corruptions (dataset shift), aspects of outlier detection gain relevancy. In this work, we show that adversarially trained autoencoders (ATA) representative of autoencoder-based outlier detection methods, yield tremendous robustness improvements over traditional neural network methods such as multi-layer perceptrons (MLP) and common ensemble methods, while maintaining a competitive classification performance. In contrast, our results also reveal that deep learning methods solely optimized for classification, tend to fail completely when exposed to dataset shift.
Author(s)
Lübbering, Max
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS
Gebauer, Michael
TU Berlin
Ramamurthy, Rajkumar
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS
Pielka, Maren
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS
Bauckhage, Christian
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS
Sifa, Rafet
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS
Hauptwerk
LWDA 2021 Workshops: FGWM, KDML, FGWI-BIA, and FGIR. Proceedings. Online resource
Konferenz
Conference "Lernen, Wissen, Daten, Analysen" (LWDA) 2021
File(s)
N-645013.pdf (889.17 KB)
Language
English
google-scholar
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS
Tags
  • dataset shift

  • representation learni...

  • outlier detection

  • One-vs-rest classific...

  • Cookie settings
  • Imprint
  • Privacy policy
  • Api
  • Send Feedback
© 2022