Adversarial robustness is one of the most challenging problems in Deep Learning and Computer Vision research. All the state-of-the-art techniques require a time-consuming procedure that creates cleverly perturbed images. Due to its cost, many solutions have been proposed to avoid Adversarial Training. However, all these attempts proved ineffective as the attacker manages to exploit spurious correlations among pixels to trigger brittle features implicitly learned by the model. This paper first introduces a new image filtering scheme called Image-Graph Extractor (IGE) that extracts the fundamental nodes of an image and their connections through a graph structure. By leveraging the IGE representation, we build a new defense method, Filtering As a Defense, that does not allow the attacker to entangle pixels to create malicious patterns. Moreover, we show that data augmentation with filtered images effectively improves the model's robustness to data corruption. We validate our techniques on CIFAR-10, CIFAR-100, and ImageNet.
翻译:反versarial 强力是深层学习和计算机视觉研究中最具挑战性的问题之一。 所有最先进的技术都需要一个费时的程序来创造巧妙的扰动图像。 由于成本高,我们提出了许多避免反向培训的解决方案。 然而,所有这些尝试都证明是无效的,因为攻击者设法利用像素之间的虚假关联来触发模型隐含的微小特征。本文首先引入了一个新的图像过滤方案,称为图像-格提取器(IGE),通过图形结构提取图像及其连接的基本节点。我们通过利用IGE代表,建立了一种新的防御方法,即“过滤作为防御”,不允许攻击者将像素连在一起制造恶意模式。此外,我们显示,通过过滤图像来增强数据,有效地提高了模型对数据腐败的稳健性。我们验证了我们在 CIRFAR-10、CIFAR-100和图像网络上的技术。