Conditional Validity of Inductive Conformal Predictors

Vladimir Vovk
Proceedings of the Asian Conference on Machine Learning, PMLR 25:475-490, 2012.

Abstract

Conformal predictors are set predictors that are automatically valid in the sense of having coverage probability equal to or exceeding a given confidence level. Inductive conformal predictors are a computationally efficient version of conformal predictors satisfying the same property of validity. However, inductive conformal predictors have been only known to control unconditional coverage probability. This paper explores various versions of conditional validity and various ways to achieve them using inductive conformal predictors and their modifications.

Cite this Paper

BibTeX
@InProceedings{pmlr-v25-vovk12, title = {Conditional Validity of Inductive Conformal Predictors}, author = {Vovk, Vladimir}, booktitle = {Proceedings of the Asian Conference on Machine Learning}, pages = {475--490}, year = {2012}, editor = {Hoi, Steven C. H. and Buntine, Wray}, volume = {25}, series = {Proceedings of Machine Learning Research}, address = {Singapore Management University, Singapore}, month = {04--06 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v25/vovk12/vovk12.pdf}, url = {https://proceedings.mlr.press/v25/vovk12.html}, abstract = {Conformal predictors are set predictors that are automatically valid in the sense of having coverage probability equal to or exceeding a given confidence level. Inductive conformal predictors are a computationally efficient version of conformal predictors satisfying the same property of validity. However, inductive conformal predictors have been only known to control unconditional coverage probability. This paper explores various versions of conditional validity and various ways to achieve them using inductive conformal predictors and their modifications.} }
Endnote
%0 Conference Paper %T Conditional Validity of Inductive Conformal Predictors %A Vladimir Vovk %B Proceedings of the Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2012 %E Steven C. H. Hoi %E Wray Buntine %F pmlr-v25-vovk12 %I PMLR %P 475--490 %U https://proceedings.mlr.press/v25/vovk12.html %V 25 %X Conformal predictors are set predictors that are automatically valid in the sense of having coverage probability equal to or exceeding a given confidence level. Inductive conformal predictors are a computationally efficient version of conformal predictors satisfying the same property of validity. However, inductive conformal predictors have been only known to control unconditional coverage probability. This paper explores various versions of conditional validity and various ways to achieve them using inductive conformal predictors and their modifications.
RIS
TY - CPAPER TI - Conditional Validity of Inductive Conformal Predictors AU - Vladimir Vovk BT - Proceedings of the Asian Conference on Machine Learning DA - 2012/11/17 ED - Steven C. H. Hoi ED - Wray Buntine ID - pmlr-v25-vovk12 PB - PMLR DP - Proceedings of Machine Learning Research VL - 25 SP - 475 EP - 490 L1 - http://proceedings.mlr.press/v25/vovk12/vovk12.pdf UR - https://proceedings.mlr.press/v25/vovk12.html AB - Conformal predictors are set predictors that are automatically valid in the sense of having coverage probability equal to or exceeding a given confidence level. Inductive conformal predictors are a computationally efficient version of conformal predictors satisfying the same property of validity. However, inductive conformal predictors have been only known to control unconditional coverage probability. This paper explores various versions of conditional validity and various ways to achieve them using inductive conformal predictors and their modifications. ER -
APA
Vovk, V.. (2012). Conditional Validity of Inductive Conformal Predictors. Proceedings of the Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 25:475-490 Available from https://proceedings.mlr.press/v25/vovk12.html.

Related Material