Rethinking Invariance Regularization in Adversarial Training to Improve Robustness-Accuracy Trade-off

arXiv:2402.14648v1 Announce Type: new
Abstract: Although adversarial training has been the state-of-the-art approach to defend against adversarial examples (AEs), they suffer from a robustness-accuracy trade-off. In this work, we revisit representation-based invariance regularization to learn discriminative yet adversarially invariant representations, aiming to mitigate this trade-off. We empirically identify two key issues hindering invariance regularization: (1) a "gradient conflict" between invariance loss and classification objectives, indicating the existence of "collapsing solutions," and (2) the mixture distribution problem arising from diverged distributions …

abstract accuracy adversarial adversarial examples adversarial training art arxiv cs.ai cs.lg examples identify key learn regularization representation robustness state trade trade-off training type work