<ul><li>Researchers propose a new class of optimal-transport-regularized divergences, D^c, to enhance the adversarial robustness of deep learning models.</li><li>The proposed ARMOR_D methods minimize the maximum expected loss over a D^c-neighborhood of the training data's empirical distribution.</li><li>ARMOR_D allows transportation and re-weighting of adversarial samples, providing enhanced adversarial re-weighting on top of adversarial sample transport.</li><li>The method demonstrates improved performance on CIFAR-10 and CIFAR-100 image recognition, outperforming existing methods against adversarial attacks.</li></ul>

Adversarially Robust Learning with Optimal Transport Regularized Divergences

Discover more