• English
  • Deutsch
  • Log In
    Password Login
    or
  • Research Outputs
  • Projects
  • Researchers
  • Institutes
  • Statistics
Repository logo
Fraunhofer-Gesellschaft
  1. Home
  2. Fraunhofer-Gesellschaft
  3. Artikel
  4. Wasserstein Dropout
 
  • Details
  • Full
Options
08 September 2022
Journal Article
Titel

Wasserstein Dropout

Abstract
Despite of its importance for safe machine learning, uncertainty quantification for neural networks is far from being solved. State-of-the-art approaches to estimate neural uncertainties are often hybrid, combining parametric models with explicit or implicit (dropout-based) ensembling. We take another pathway and propose a novel approach to uncertainty quantification for regression tasks, Wasserstein dropout, that is purely non-parametric. Technically, it captures aleatoric uncertainty by means of dropout-based sub-network distributions. This is accomplished by a new objective which minimizes the Wasserstein distance between the label distribution and the model distribution. An extensive empirical analysis shows that Wasserstein dropout outperforms state-of-the-art methods, on vanilla test data as well as under distributional shift in terms of producing more accurate and stable uncertainty estimates.
Author(s)
Sicking, Joachim
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS
Akila, Maram
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS
Pintz, Maximilian Alexander
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS
Wirtz, Tim
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS
Wrobel, Stefan
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS
Fischer, Asja
Ruhr-Univ. Bochum
Zeitschrift
Machine learning
Project(s)
KI Absicherung
ML2R
EXC 2092: Cyber-Sicherheit im Zeitalter großskaliger Angreifer
Funder
Bundesministerium für Wirtschaft und Klimaschutz BMWK
Bundesministerium für Bildung und Forschung -BMBF-
Deutsche Forschungsgemeinschaft -DFG-, Bonn
Thumbnail Image
DOI
10.1007/s10994-022-06230-8
Language
English
google-scholar
Fraunhofer-Institut für Intelligente Analyse- und Informationssysteme IAIS
Tags
  • Safe Machine Learning...

  • Regression Neural Net...

  • Uncertainty Estimatio...

  • Aleatoric Uncertainty...

  • Dropout

  • Object Detection

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