Paper published in a journal (Scientific congresses and symposiums)
Learning to Pivot with Adversarial Networks
Louppe, Gilles; Kagan, Michael; Cranmer, Kyle
2016In Advances in Neural Information Processing Systems, 30
Peer Reviewed verified by ORBi
 

Files


Full Text
6699-learning-to-pivot-with-adversarial-networks.pdf
Publisher postprint (532.82 kB)
Download

All documents in ORBi are protected by a user license.

Send to



Details



Keywords :
Statistics - Machine Learning; Computer Science - Learning; Computer Science - Neural and Evolutionary Computing; Physics - Data Analysis; Statistics and Probability; Statistics - Methodology
Abstract :
[en] Several techniques for domain adaptation have been proposed to account for differences in the distribution of the data used for training and testing. The majority of this work focuses on a binary domain label. Similar problems occur in a scientific context where there may be a continuous family of plausible data generation processes associated to the presence of systematic uncertainties. Robust inference is possible if it is based on a pivot -- a quantity whose distribution does not depend on the unknown values of the nuisance parameters that parametrize this family of data generation processes. In this work, we introduce and derive theoretical results for a training procedure based on adversarial networks for enforcing the pivotal property (or, equivalently, fairness with respect to continuous attributes) on a predictive model. The method includes a hyperparameter to control the trade-off between accuracy and robustness. We demonstrate the effectiveness of this approach with a toy example and examples from particle physics.
Disciplines :
Computer science
Author, co-author :
Louppe, Gilles  ;  Université de Liège - ULiège > Dép. d'électric., électron. et informat. (Inst.Montefiore) > Big Data
Kagan, Michael
Cranmer, Kyle
Language :
English
Title :
Learning to Pivot with Adversarial Networks
Publication date :
November 2016
Event name :
Neural Information Processing Systems Conference (NIPS) 2017
Event place :
Long Beach, United States
Event date :
December 2017
Audience :
International
Journal title :
Advances in Neural Information Processing Systems
ISSN :
1049-5258
Publisher :
Morgan Kaufmann Publishers, San Mateo, United States - California
Volume :
30
Peer reviewed :
Peer Reviewed verified by ORBi
Available on ORBi :
since 28 June 2018

Statistics


Number of views
40 (5 by ULiège)
Number of downloads
18 (1 by ULiège)

Scopus citations®
 
79
Scopus citations®
without self-citations
75

Bibliography


Similar publications



Contact ORBi