Conformal predictive decision making

Vladimir Vovk, Claus Bendtsen
Proceedings of the Seventh Workshop on Conformal and Probabilistic Prediction and Applications, PMLR 91:52-62, 2018.

Abstract

This note explains how conformal predictive distributions can be used for the purpose of decision-making. Namely, a major limitation of conformal predictive distributions is that, at this time, they are only applicable to regression problems, where the label is a real number; however, this does not prevent them from being used in a general problem of decision making. The resulting methodology of conformal predictive decision making is illustrated on a small benchmark data set. Our main theoretical observation is that there exists an asymptotically efficient predictive decision-making system which can be obtained by using our methodology (and therefore, satisfying the standard property of validity).

Cite this Paper

BibTeX
@InProceedings{pmlr-v91-vovk18b, title = {Conformal predictive decision making}, author = {Vovk, Vladimir and Bendtsen, Claus}, booktitle = {Proceedings of the Seventh Workshop on Conformal and Probabilistic Prediction and Applications}, pages = {52--62}, year = {2018}, editor = {Gammerman, Alex and Vovk, Vladimir and Luo, Zhiyuan and Smirnov, Evgueni and Peeters, Ralf}, volume = {91}, series = {Proceedings of Machine Learning Research}, month = {11--13 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v91/vovk18b/vovk18b.pdf}, url = {https://proceedings.mlr.press/v91/vovk18b.html}, abstract = {This note explains how conformal predictive distributions can be used for the purpose of decision-making. Namely, a major limitation of conformal predictive distributions is that, at this time, they are only applicable to regression problems, where the label is a real number; however, this does not prevent them from being used in a general problem of decision making. The resulting methodology of conformal predictive decision making is illustrated on a small benchmark data set. Our main theoretical observation is that there exists an asymptotically efficient predictive decision-making system which can be obtained by using our methodology (and therefore, satisfying the standard property of validity).} }
Endnote
%0 Conference Paper %T Conformal predictive decision making %A Vladimir Vovk %A Claus Bendtsen %B Proceedings of the Seventh Workshop on Conformal and Probabilistic Prediction and Applications %C Proceedings of Machine Learning Research %D 2018 %E Alex Gammerman %E Vladimir Vovk %E Zhiyuan Luo %E Evgueni Smirnov %E Ralf Peeters %F pmlr-v91-vovk18b %I PMLR %P 52--62 %U https://proceedings.mlr.press/v91/vovk18b.html %V 91 %X This note explains how conformal predictive distributions can be used for the purpose of decision-making. Namely, a major limitation of conformal predictive distributions is that, at this time, they are only applicable to regression problems, where the label is a real number; however, this does not prevent them from being used in a general problem of decision making. The resulting methodology of conformal predictive decision making is illustrated on a small benchmark data set. Our main theoretical observation is that there exists an asymptotically efficient predictive decision-making system which can be obtained by using our methodology (and therefore, satisfying the standard property of validity).
APA
Vovk, V. & Bendtsen, C.. (2018). Conformal predictive decision making. Proceedings of the Seventh Workshop on Conformal and Probabilistic Prediction and Applications, in Proceedings of Machine Learning Research 91:52-62 Available from https://proceedings.mlr.press/v91/vovk18b.html.

Related Material