<ul><li>Dataset distillation (DD) allows datasets to be distilled to fractions of their original size while preserving the rich distributional information so that models trained on the distilled datasets can achieve a comparable accuracy while saving significant computational loads.</li><li>This paper explores a new perspective of dataset distillation by embedding adversarial robustness, enabling models trained on these datasets to maintain high accuracy and better adversarial robustness.</li><li>The proposed method incorporates curvature regularization into the distillation process, resulting in improved accuracy and robustness compared to standard adversarial training, with lower computational overhead.</li><li>Empirical experiments demonstrate that the method generates robust distilled datasets capable of withstanding various adversarial attacks.</li></ul>

Towards Adversarially Robust Dataset Distillation by Curvature Regularization

Discover more