Conformal Prediction under Hypergraphical Models

Abstract : Conformal predictors are usually defined and studied under the exchangeability assumption. However, their definition can be extended to a wide class of statistical models, called online compression models, while retaining their property of automatic validity. This paper is devoted to conformal prediction under hypergraphical models that are more specific than the exchangeability model. Namely, we define two natural classes of conformity measures for such hypergraphical models and study the corresponding conformal predictors empirically on benchmark LED data sets. Our experiments show that they are more efficient than conformal predictors that use only the exchangeability assumption.
Document type :
Conference papers
Complete list of metadatas

Cited literature [6 references]  Display  Hide  Download

https://hal.inria.fr/hal-01459632
Contributor : Hal Ifip <>
Submitted on : Tuesday, February 7, 2017 - 1:05:42 PM
Last modification on : Friday, December 1, 2017 - 1:16:34 AM
Long-term archiving on : Monday, May 8, 2017 - 2:15:25 PM

File

978-3-642-41142-7_38_Chapter.p...
Files produced by the author(s)

Licence


Distributed under a Creative Commons Attribution 4.0 International License

Identifiers

Citation

Valentina Fedorova, Alex Gammerman, Ilia Nouretdinov, Vladimir Vovk. Conformal Prediction under Hypergraphical Models. 9th Artificial Intelligence Applications and Innovations (AIAI), Sep 2013, Paphos, Greece. pp.371-383, ⟨10.1007/978-3-642-41142-7_38⟩. ⟨hal-01459632⟩

Share

Metrics

Record views

413

Files downloads

91