A Curious Case of Remarkable Resilience to Gradient Attacks via Fully Convolutional and Differentiable Front End with a Skip Connection

arXiv:2402.17018v1 Announce Type: new
Abstract: We tested front-end enhanced neural models where a frozen classifier was prepended by a differentiable and fully convolutional model with a skip connection. By training them using a small learning rate for about one epoch, we obtained models that retained the accuracy of the backbone classifier while being unusually resistant to gradient attacks including APGD and FAB-T attacks from the AutoAttack package, which we attributed to gradient masking. The gradient masking phenomenon is not new, …

abstract arxiv attacks case classifier cs.ai cs.cv cs.lg differentiable front-end gradient rate resilience small them training type via