GTAT: Adversarial Training with Generated Triplets

Baoli Wang, Xinxin Fan, Quanliang Jing, Yueyang Su, Jingwei Li, Jingping Bi

Abstract: To circumvent the grave problem of present adversarial training methods, i.e. distortion of classification surface, we in this paper propose a generated Triplet-based adversarial training method-GTAT, in which a Generator generates a semi-hard Triplet by design, rather than directly invoking the existing clean examples and adversarial examples. Through this kind of generated semi-hard Triplet constraint, GTAT can reshape the classification boundaries appropriately across various classes, arising from two-facet synergies: i) pull the intra-class examples together with tight distances; and ii) push away the inter-class examples with broad distances. This synergy will simplify and broaden the classification surfaces across different classes. Extensive experiments on the popular MNIST and CIFAR-10 datasets show that our proposed GTAT significantly outperforms other state-of-the-art adversarial training methods. We believe GTAT opens a door for the adversarial training from a new horizon of rationally generating semi-hard Triplet-satisfied adversarial training (retraining) examples, instead of straightly performing retraining on the generated adversarial examples and existing clean examples, or on the generated adversarial examples only.

External IDs:dblp:conf/ijcnn/WangFJSLB22