Details

Autor(en) / Beteiligte

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

Titel

Robust Importance-Weighted Cross-Validation Under Sample Selection Bias

Ist Teil von

• 2019 IEEE 29th International Workshop on Machine Learning for Signal Processing (MLSP), 2019, p.1-6

Ort / Verlag

IEEE

Erscheinungsjahr

2019

Quelle

IEEE Xplore

Beschreibungen/Notizen

• 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.