Coreset-based Conformal Prediction for Large-scale Learning

Nery Riquelme-Granada, Khuong Nguyen, Zhiyuan Luo
; Proceedings of the Eighth Symposium on Conformal and Probabilistic Prediction and Applications, PMLR 105:142-162, 2019.

Abstract

As the volume of data increase rapidly, most traditional machine learning algorithms become computationally prohibitive. Furthermore, the available data can be so big that a single machine’s memory can easily be overflown. We propose Coreset-Based Conformal Prediction, a strategy for dealing with big data by applying conformal predictors to a weighted summary of data—namely the coreset. We compare our approach against standalone inductive conformal predictors over three large competition-grade datasets to demonstrate that our coreset-based strategy may not only significantly improve the learning speed, but also retains predictions validity and the predictors’ efficiency.

Cite this Paper


BibTeX
@InProceedings{pmlr-v105-riquelme-granada19a, title = {Coreset-based Conformal Prediction for Large-scale Learning}, author = {Riquelme-Granada, Nery and Nguyen, Khuong and Luo, Zhiyuan}, booktitle = {Proceedings of the Eighth Symposium on Conformal and Probabilistic Prediction and Applications}, pages = {142--162}, year = {2019}, editor = {Alex Gammerman and Vladimir Vovk and Zhiyuan Luo and Evgueni Smirnov}, volume = {105}, series = {Proceedings of Machine Learning Research}, address = {Golden Sands, Bulgaria}, month = {09--11 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v105/riquelme-granada19a/riquelme-granada19a.pdf}, url = {http://proceedings.mlr.press/v105/riquelme-granada19a.html}, abstract = {As the volume of data increase rapidly, most traditional machine learning algorithms become computationally prohibitive. Furthermore, the available data can be so big that a single machine’s memory can easily be overflown. We propose Coreset-Based Conformal Prediction, a strategy for dealing with big data by applying conformal predictors to a weighted summary of data—namely the coreset. We compare our approach against standalone inductive conformal predictors over three large competition-grade datasets to demonstrate that our coreset-based strategy may not only significantly improve the learning speed, but also retains predictions validity and the predictors’ efficiency.} }
Endnote
%0 Conference Paper %T Coreset-based Conformal Prediction for Large-scale Learning %A Nery Riquelme-Granada %A Khuong Nguyen %A Zhiyuan Luo %B Proceedings of the Eighth Symposium on Conformal and Probabilistic Prediction and Applications %C Proceedings of Machine Learning Research %D 2019 %E Alex Gammerman %E Vladimir Vovk %E Zhiyuan Luo %E Evgueni Smirnov %F pmlr-v105-riquelme-granada19a %I PMLR %J Proceedings of Machine Learning Research %P 142--162 %U http://proceedings.mlr.press %V 105 %W PMLR %X As the volume of data increase rapidly, most traditional machine learning algorithms become computationally prohibitive. Furthermore, the available data can be so big that a single machine’s memory can easily be overflown. We propose Coreset-Based Conformal Prediction, a strategy for dealing with big data by applying conformal predictors to a weighted summary of data—namely the coreset. We compare our approach against standalone inductive conformal predictors over three large competition-grade datasets to demonstrate that our coreset-based strategy may not only significantly improve the learning speed, but also retains predictions validity and the predictors’ efficiency.
APA
Riquelme-Granada, N., Nguyen, K. & Luo, Z.. (2019). Coreset-based Conformal Prediction for Large-scale Learning. Proceedings of the Eighth Symposium on Conformal and Probabilistic Prediction and Applications, in PMLR 105:142-162

Related Material