Robust importance-weighted cross-validation under sample selection bias

Wouter M. Kouw, Jesse H. Krijthe, Marco Loog

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)
38 Downloads (Pure)

Abstract

Cross-validation under sample selection bias can, in principle, be done by importance-weighting the empirical risk. However, the importance-weighted risk estimator produces suboptimal hyperparameter estimates in problem settings where large weights arise with high probability. We study its sampling variance as a function of the training data distribution and introduce a control variate to increase its robustness to problematically large weights.
Original languageEnglish
Title of host publication2019 IEEE 29th International Workshop on Machine Learning for Signal Processing, MLSP 2019
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
Number of pages6
ISBN (Electronic)978-1-7281-0824-7
DOIs
Publication statusPublished - 5 Dec 2019
EventIEEE International Workshop on Machine Learning for Signal Processing, MLSP 2019 - University of Pittsburgh, Pittsburgh, United States
Duration: 13 Oct 201916 Dec 2019
Conference number: 29
https://www.ieeemlsp.cc/

Conference

ConferenceIEEE International Workshop on Machine Learning for Signal Processing, MLSP 2019
Abbreviated titleMLSP 2019
Country/TerritoryUnited States
CityPittsburgh
Period13/10/1916/12/19
Internet address

Keywords

  • Sample selection bias
  • cross-validation

Fingerprint

Dive into the research topics of 'Robust importance-weighted cross-validation under sample selection bias'. Together they form a unique fingerprint.

Cite this