The knockoff filter for FDR control in group-sparse and multitask regression

Ran Dai, Rina Barber
Proceedings of The 33rd International Conference on Machine Learning, PMLR 48:1851-1859, 2016.

Abstract

We propose the group knockoff filter, a method for false discovery rate control in a linear regression setting where the features are grouped, and we would like to select a set of relevant groups which have a nonzero effect on the response. By considering the set of true and false discoveries at the group level, this method gains power relative to sparse regression methods. We also apply our method to the multitask regression problem where multiple response variables share similar sparsity patterns across the set of possible features. Empirically, the group knockoff filter successfully controls false discoveries at the group level in both settings, with substantially more discoveries made by leveraging the group structure.

Cite this Paper

BibTeX
@InProceedings{pmlr-v48-daia16, title = {The knockoff filter for FDR control in group-sparse and multitask regression}, author = {Dai, Ran and Barber, Rina}, booktitle = {Proceedings of The 33rd International Conference on Machine Learning}, pages = {1851--1859}, year = {2016}, editor = {Balcan, Maria Florina and Weinberger, Kilian Q.}, volume = {48}, series = {Proceedings of Machine Learning Research}, address = {New York, New York, USA}, month = {20--22 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v48/daia16.pdf}, url = {https://proceedings.mlr.press/v48/daia16.html}, abstract = {We propose the group knockoff filter, a method for false discovery rate control in a linear regression setting where the features are grouped, and we would like to select a set of relevant groups which have a nonzero effect on the response. By considering the set of true and false discoveries at the group level, this method gains power relative to sparse regression methods. We also apply our method to the multitask regression problem where multiple response variables share similar sparsity patterns across the set of possible features. Empirically, the group knockoff filter successfully controls false discoveries at the group level in both settings, with substantially more discoveries made by leveraging the group structure.} }
Endnote
%0 Conference Paper %T The knockoff filter for FDR control in group-sparse and multitask regression %A Ran Dai %A Rina Barber %B Proceedings of The 33rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2016 %E Maria Florina Balcan %E Kilian Q. Weinberger %F pmlr-v48-daia16 %I PMLR %P 1851--1859 %U https://proceedings.mlr.press/v48/daia16.html %V 48 %X We propose the group knockoff filter, a method for false discovery rate control in a linear regression setting where the features are grouped, and we would like to select a set of relevant groups which have a nonzero effect on the response. By considering the set of true and false discoveries at the group level, this method gains power relative to sparse regression methods. We also apply our method to the multitask regression problem where multiple response variables share similar sparsity patterns across the set of possible features. Empirically, the group knockoff filter successfully controls false discoveries at the group level in both settings, with substantially more discoveries made by leveraging the group structure.
RIS
TY - CPAPER TI - The knockoff filter for FDR control in group-sparse and multitask regression AU - Ran Dai AU - Rina Barber BT - Proceedings of The 33rd International Conference on Machine Learning DA - 2016/06/11 ED - Maria Florina Balcan ED - Kilian Q. Weinberger ID - pmlr-v48-daia16 PB - PMLR DP - Proceedings of Machine Learning Research VL - 48 SP - 1851 EP - 1859 L1 - http://proceedings.mlr.press/v48/daia16.pdf UR - https://proceedings.mlr.press/v48/daia16.html AB - We propose the group knockoff filter, a method for false discovery rate control in a linear regression setting where the features are grouped, and we would like to select a set of relevant groups which have a nonzero effect on the response. By considering the set of true and false discoveries at the group level, this method gains power relative to sparse regression methods. We also apply our method to the multitask regression problem where multiple response variables share similar sparsity patterns across the set of possible features. Empirically, the group knockoff filter successfully controls false discoveries at the group level in both settings, with substantially more discoveries made by leveraging the group structure. ER -
APA
Dai, R. & Barber, R.. (2016). The knockoff filter for FDR control in group-sparse and multitask regression. Proceedings of The 33rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 48:1851-1859 Available from https://proceedings.mlr.press/v48/daia16.html.

Related Material