1 paper
Deeksha Adil, Jarosław Błasiok
Covariate shift, a widely used assumption in tackling {\it distributional shift} (when training and test distributions differ), focuses on scenarios where the distribution of the l…