TY - CONF
AU - Hamdan, Sami
AU - Love, Bradley C.
AU - Polier, Georg von
AU - Weis, Susanne
AU - Schwender, Holger
AU - Eickhoff, Simon
AU - Patil, Kaustubh
TI - Cofound-Leakage: Confound Removal In Machine Learning Leads To Leakage
M1 - FZJ-2023-03045
PY - 2023
N1 - Acknowledgments: This work was partly supported by the Helmholtz-AI project DeGen, the Helmholtz Portfolio Theme ‘Supercomputing and Modeling for the Human Brain’ and Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) Poster: Pitfalls of Confound Regression in Machine Learning. Der Postertitel lautet anders, doch das war ok fuer OHBM Veranstalter
AB - Modern Machine Learning (ML) approaches are now regularly employed forindividual-level prediction, e.g. personalized medicine.Particularly in such critical-decision making, it is of utmost importance to not onlyachieve high accuracy but also to trust that models rely on actual features-targetrelationships [1, 2]. To this end, it is crucial to consider confounding variables as theycan obstruct the features-target relationship. For instance, a researcher might wantto identify a biomarker showing high classification accuracy between controls andpatients. However, the model might have just learned simpler confounders like ageor sex as a good proxy of the disease [3]. To counteract such unwanted confoundingeffects, investigators often use linear models to remove confounding variables fromeach feature separately before employing ML. While this confound regression (CR)approach is popular [4], its pitfalls, especially when paired with non-linear MLmodels, are not well understood.
T2 - Organization for Human Brain Mapping (OHBM)
CY - 22 Jul 2023 - 26 Jul 2023, Montreal (Canada)
Y2 - 22 Jul 2023 - 26 Jul 2023
M2 - Montreal, Canada
LB - PUB:(DE-HGF)24
DO - DOI:10.34734/FZJ-2023-03045
UR - https://juser.fz-juelich.de/record/1010405
ER -