liu.seSearch for publications in DiVA
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • 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
Calibration tests beyond classification
Department of Information Technology Uppsala University, Sweden.ORCID iD: 0000-0001-9282-053x
Linköping University, Department of Computer and Information Science, The Division of Statistics and Machine Learning. Linköping University, Faculty of Arts and Sciences.ORCID iD: 0000-0003-3749-5820
Department of Information Technology Uppsala University, Sweden.ORCID iD: 0000-0002-6698-0166
2021 (English)In: ICLR 2021 - 9th International Conference on Learning Representations Proceedings, International Conference on Learning Representations, ICLR , 2021, p. 1-37Conference paper, Published paper (Refereed)
Abstract [en]

Most supervised machine learning tasks are subject to irreducible prediction errors. Probabilistic predictive models address this limitation by providing probability distributions that represent a belief over plausible targets, rather than point estimates. Such models can be a valuable tool in decision-making under uncertainty, provided that the model output is meaningful and interpretable. Calibrated models guarantee that the probabilistic predictions are neither over- nor under-confident. In the machine learning literature, different measures and statistical tests have been proposed and studied for evaluating the calibration of classification models. For regression problems, however, research has been focused on a weaker condition of calibration based on predicted quantiles for real-valued targets. In this paper, we propose the first framework that unifies calibration evaluation and tests for probabilistic predictive models. It applies to any such model, including classification and regression models of arbitrary dimension. Furthermore, the framework generalizes existing measures and provides a more intuitive reformulation of a recently proposed framework for calibration in multi-class classification.

Place, publisher, year, edition, pages
International Conference on Learning Representations, ICLR , 2021. p. 1-37
Keywords [en]
calibration, uncertainty quantification, framework, integral probability metric, maximum mean discrepancy
National Category
Probability Theory and Statistics
Research subject
Mathematical Statistics
Identifiers
URN: urn:nbn:se:liu:diva-188940Scopus ID: 2-s2.0-85147937089OAI: oai:DiVA.org:liu-188940DiVA, id: diva2:1700801
Conference
International Conference on Learning Representations, Virtual conference, May 3 - May 7, 2021
Available from: 2020-12-23 Created: 2022-10-03 Last updated: 2024-08-23Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

ScopusLink to paperVideo presentationSource code of the experiments

Authority records

Widmann, DavidLindsten, FredrikZachariah, Dave

Search in DiVA

By author/editor
Widmann, DavidLindsten, FredrikZachariah, Dave
By organisation
The Division of Statistics and Machine LearningFaculty of Arts and Sciences
Probability Theory and Statistics

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 47 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • 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