Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

**Robust Importance-Weighted Cross-Validation under Sample Selection Bias.** / Kouw, Wouter M.; Krijthe, Jesse H.; Loog, Marco.

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

Kouw, WM, Krijthe, JH & Loog, M 2019, Robust Importance-Weighted Cross-Validation under Sample Selection Bias. in *2019 IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP).*, 8918731, IEEE, Piscataway, pp. 1-6, 29th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2019, Pittsburgh, United States, 13/10/19. https://doi.org/10.1109/MLSP.2019.8918731

Kouw, W. M., Krijthe, J. H., & Loog, M. (2019). Robust Importance-Weighted Cross-Validation under Sample Selection Bias. In *2019 IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP) *(pp. 1-6). [8918731] IEEE. https://doi.org/10.1109/MLSP.2019.8918731

Kouw WM, Krijthe JH, Loog M. Robust Importance-Weighted Cross-Validation under Sample Selection Bias. In 2019 IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP). Piscataway: IEEE. 2019. p. 1-6. 8918731 https://doi.org/10.1109/MLSP.2019.8918731

@inproceedings{da88aa2f659e44e9a9c39c1c625afdd7,

title = "Robust Importance-Weighted Cross-Validation under Sample Selection Bias",

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.",

keywords = "cross-validation, Sample selection bias",

author = "Kouw, {Wouter M.} and Krijthe, {Jesse H.} and Marco Loog",

year = "2019",

doi = "10.1109/MLSP.2019.8918731",

language = "English",

isbn = "978-1-7281-0825-4",

pages = "1--6",

booktitle = "2019 IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP)",

publisher = "IEEE",

address = "United States",

note = "29th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2019 ; Conference date: 13-10-2019 Through 16-10-2019",

}

TY - GEN

T1 - Robust Importance-Weighted Cross-Validation under Sample Selection Bias

AU - Kouw, Wouter M.

AU - Krijthe, Jesse H.

AU - Loog, Marco

PY - 2019

Y1 - 2019

N2 - 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.

AB - 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.

KW - cross-validation

KW - Sample selection bias

UR - http://www.scopus.com/inward/record.url?scp=85077676600&partnerID=8YFLogxK

U2 - 10.1109/MLSP.2019.8918731

DO - 10.1109/MLSP.2019.8918731

M3 - Conference contribution

AN - SCOPUS:85077676600

SN - 978-1-7281-0825-4

SP - 1

EP - 6

BT - 2019 IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP)

PB - IEEE

CY - Piscataway

T2 - 29th IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2019

Y2 - 13 October 2019 through 16 October 2019

ER -

ID: 68748044