HARD: Hard Augmentations for Robust Distillation

no code implementations • 24 May 2023 • Arne F. Nix, Max F. Burg, Fabian H. Sinz

To improve these aspects of KD, we propose Hard Augmentations for Robust Distillation (HARD), a generally applicable data augmentation framework, that generates synthetic data points for which the teacher and the student disagree.

Data Augmentation Domain Generalization +1