Learning robust visual representations using data augmentation invariance

OOD
Re-assign community