Conformal multi-target regression using neural networks

Soundouss Messoudi, Sébastien Destercke, Sylvain Rousseau
Proceedings of the Ninth Symposium on Conformal and Probabilistic Prediction and Applications, PMLR 128:65-83, 2020.

Abstract

Multi-task learning is a domain that is still not fully studied in the conformal prediction framework, and this is particularly true for multi-target regression. Our work uses inductive conformal prediction along with deep neural networks to handle multi-target regression by exploring multiple extensions of existing single-target non-conformity measures and proposing new ones. This paper presents our approaches to work with conformal prediction in the multiple regression setting, as well as the results of our conducted experiments.

Cite this Paper

BibTeX
@InProceedings{pmlr-v128-messoudi20a, title = {Conformal multi-target regression using neural networks}, author = {Messoudi, Soundouss and Destercke, S\'{e}bastien and Rousseau, Sylvain}, booktitle = {Proceedings of the Ninth Symposium on Conformal and Probabilistic Prediction and Applications}, pages = {65--83}, year = {2020}, editor = {Gammerman, Alexander and Vovk, Vladimir and Luo, Zhiyuan and Smirnov, Evgueni and Cherubin, Giovanni}, volume = {128}, series = {Proceedings of Machine Learning Research}, month = {09--11 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v128/messoudi20a/messoudi20a.pdf}, url = {https://proceedings.mlr.press/v128/messoudi20a.html}, abstract = {Multi-task learning is a domain that is still not fully studied in the conformal prediction framework, and this is particularly true for multi-target regression. Our work uses inductive conformal prediction along with deep neural networks to handle multi-target regression by exploring multiple extensions of existing single-target non-conformity measures and proposing new ones. This paper presents our approaches to work with conformal prediction in the multiple regression setting, as well as the results of our conducted experiments.} }
Endnote
%0 Conference Paper %T Conformal multi-target regression using neural networks %A Soundouss Messoudi %A Sébastien Destercke %A Sylvain Rousseau %B Proceedings of the Ninth Symposium on Conformal and Probabilistic Prediction and Applications %C Proceedings of Machine Learning Research %D 2020 %E Alexander Gammerman %E Vladimir Vovk %E Zhiyuan Luo %E Evgueni Smirnov %E Giovanni Cherubin %F pmlr-v128-messoudi20a %I PMLR %P 65--83 %U https://proceedings.mlr.press/v128/messoudi20a.html %V 128 %X Multi-task learning is a domain that is still not fully studied in the conformal prediction framework, and this is particularly true for multi-target regression. Our work uses inductive conformal prediction along with deep neural networks to handle multi-target regression by exploring multiple extensions of existing single-target non-conformity measures and proposing new ones. This paper presents our approaches to work with conformal prediction in the multiple regression setting, as well as the results of our conducted experiments.
APA
Messoudi, S., Destercke, S. & Rousseau, S.. (2020). Conformal multi-target regression using neural networks. Proceedings of the Ninth Symposium on Conformal and Probabilistic Prediction and Applications, in Proceedings of Machine Learning Research 128:65-83 Available from https://proceedings.mlr.press/v128/messoudi20a.html.

Related Material