Aggregation of Multiple Knockoffs

Tuan-Binh Nguyen, Jerome-Alexis Chevalier, Bertrand Thirion, Sylvain Arlot
Proceedings of the 37th International Conference on Machine Learning, PMLR 119:7283-7293, 2020.

Abstract

We develop an extension of the knockoff inference procedure, introduced by Barber & Candes (2015). This new method, called Aggregation of Multiple Knockoffs (AKO), addresses the instability inherent to the random nature of knockoff-based inference. Specifically, AKO improves both the stability and power compared with the original knockoff algorithm while still maintaining guarantees for false discovery rate control. We provide a new inference procedure, prove its core properties, and demonstrate its benefits in a set of experiments on synthetic and real datasets.

Cite this Paper


BibTeX
@InProceedings{pmlr-v119-nguyen20a, title = {Aggregation of Multiple Knockoffs}, author = {Nguyen, Tuan-Binh and Chevalier, Jerome-Alexis and Thirion, Bertrand and Arlot, Sylvain}, booktitle = {Proceedings of the 37th International Conference on Machine Learning}, pages = {7283--7293}, year = {2020}, editor = {III, Hal Daumé and Singh, Aarti}, volume = {119}, series = {Proceedings of Machine Learning Research}, month = {13--18 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v119/nguyen20a/nguyen20a.pdf}, url = {https://proceedings.mlr.press/v119/nguyen20a.html}, abstract = {We develop an extension of the knockoff inference procedure, introduced by Barber & Candes (2015). This new method, called Aggregation of Multiple Knockoffs (AKO), addresses the instability inherent to the random nature of knockoff-based inference. Specifically, AKO improves both the stability and power compared with the original knockoff algorithm while still maintaining guarantees for false discovery rate control. We provide a new inference procedure, prove its core properties, and demonstrate its benefits in a set of experiments on synthetic and real datasets.} }
Endnote
%0 Conference Paper %T Aggregation of Multiple Knockoffs %A Tuan-Binh Nguyen %A Jerome-Alexis Chevalier %A Bertrand Thirion %A Sylvain Arlot %B Proceedings of the 37th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2020 %E Hal Daumé III %E Aarti Singh %F pmlr-v119-nguyen20a %I PMLR %P 7283--7293 %U https://proceedings.mlr.press/v119/nguyen20a.html %V 119 %X We develop an extension of the knockoff inference procedure, introduced by Barber & Candes (2015). This new method, called Aggregation of Multiple Knockoffs (AKO), addresses the instability inherent to the random nature of knockoff-based inference. Specifically, AKO improves both the stability and power compared with the original knockoff algorithm while still maintaining guarantees for false discovery rate control. We provide a new inference procedure, prove its core properties, and demonstrate its benefits in a set of experiments on synthetic and real datasets.
APA
Nguyen, T., Chevalier, J., Thirion, B. & Arlot, S.. (2020). Aggregation of Multiple Knockoffs. Proceedings of the 37th International Conference on Machine Learning, in Proceedings of Machine Learning Research 119:7283-7293 Available from https://proceedings.mlr.press/v119/nguyen20a.html.

Related Material