Multi-class probabilistic classification using inductive and cross Venn–Abers predictors

Valery Manokhin
Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications, PMLR 60:228-240, 2017.

Abstract

Inductive (IVAP) and cross (CVAP) Venn–Abers predictors are computationally efficient algorithms for probabilistic prediction in binary classification problems. We present a new approach to multi-class probability estimation by turning IVAPs and CVAPs into multi-class probabilistic predictors. The proposed multi-class predictors are experimentally more accurate than both uncalibrated predictors and existing calibration methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v60-manokhin17a, title = {Multi-class probabilistic classification using inductive and cross {V}enn–{A}bers predictors}, author = {Manokhin, Valery}, booktitle = {Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications}, pages = {228--240}, year = {2017}, editor = {Gammerman, Alex and Vovk, Vladimir and Luo, Zhiyuan and Papadopoulos, Harris}, volume = {60}, series = {Proceedings of Machine Learning Research}, month = {13--16 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v60/manokhin17a/manokhin17a.pdf}, url = {https://proceedings.mlr.press/v60/manokhin17a.html}, abstract = {Inductive (IVAP) and cross (CVAP) Venn–Abers predictors are computationally efficient algorithms for probabilistic prediction in binary classification problems. We present a new approach to multi-class probability estimation by turning IVAPs and CVAPs into multi-class probabilistic predictors. The proposed multi-class predictors are experimentally more accurate than both uncalibrated predictors and existing calibration methods.} }
Endnote
%0 Conference Paper %T Multi-class probabilistic classification using inductive and cross Venn–Abers predictors %A Valery Manokhin %B Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications %C Proceedings of Machine Learning Research %D 2017 %E Alex Gammerman %E Vladimir Vovk %E Zhiyuan Luo %E Harris Papadopoulos %F pmlr-v60-manokhin17a %I PMLR %P 228--240 %U https://proceedings.mlr.press/v60/manokhin17a.html %V 60 %X Inductive (IVAP) and cross (CVAP) Venn–Abers predictors are computationally efficient algorithms for probabilistic prediction in binary classification problems. We present a new approach to multi-class probability estimation by turning IVAPs and CVAPs into multi-class probabilistic predictors. The proposed multi-class predictors are experimentally more accurate than both uncalibrated predictors and existing calibration methods.
APA
Manokhin, V.. (2017). Multi-class probabilistic classification using inductive and cross Venn–Abers predictors. Proceedings of the Sixth Workshop on Conformal and Probabilistic Prediction and Applications, in Proceedings of Machine Learning Research 60:228-240 Available from https://proceedings.mlr.press/v60/manokhin17a.html.

Related Material