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
Handling Small Calibration Sets in Mondrian Inductive Conformal Regressors
Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
Show others and affiliations
2015 (English)In: Statistical Learning and Data Sciences: Third International Symposium, SLDS 2015 Egham, UK, April 20–23, 2015 Proceedings / [ed] Alexander Gammerman, Vladimir Vovk, Harris Papadopoulos, Cham: Springer, 2015, 271-280 p.Conference paper, Published paper (Refereed)
Resource type
Text
Abstract [en]

In inductive conformal prediction, calibration sets must contain an adequate number of instances to support the chosen confidence level. This problem is particularly prevalent when using Mondrian inductive conformal prediction, where the input space is partitioned into independently valid prediction regions. In this study, Mondrian conformal regressors, in the form of regression trees, are used to investigate two problematic aspects of small calibration sets. If there are too few calibration instances to support the significance level, we suggest using either extrapolation or altering the model. In situations where the desired significance level is between two calibration instances, the standard procedure is to choose the more nonconforming one, thus guaranteeing validity, but producing conservative conformal predictors. The suggested solution is to use interpolation between calibration instances. All proposed techniques are empirically evaluated and compared to the standard approach on 30 benchmark data set . The results show that while extrapolation often results in invalid models, interpolation works extremely well and provides increased efficiency with preserved empirical validity.

Place, publisher, year, edition, pages
Cham: Springer, 2015. 271-280 p.
Series
Lecture Notes in Computer Science, ISSN 0302-9743 ; 9047
National Category
Information Systems
Research subject
Computer and Systems Sciences
Identifiers
URN: urn:nbn:se:su:diva-124665ISBN: 978-3-319-17090-9 (print)OAI: oai:DiVA.org:su-124665DiVA: diva2:890500
Conference
Third International Symposium, SLDS 2015, Egham, UK, April 20–23, 2015
Available from: 2016-01-04 Created: 2016-01-04 Last updated: 2016-11-22Bibliographically approved

Open Access in DiVA

No full text

Search in DiVA

By author/editor
Boström, Henrik
By organisation
Department of Computer and Systems Sciences
Information Systems

Search outside of DiVA

GoogleGoogle Scholar

Total: 29 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