A range of defense methods have been proposed to improve the robustness of neural networks on adversarial examples, among which provable defense methods have been demonstrated to be effective to train neural networks that are certifiably robust to the attacker. However, most of these provable defense methods treat all examples equally during training process, which ignore the inconsistent constraint of certified robustness between correctly classified (natural) and misclassified examples. In this paper, we explore this inconsistency caused by misclassified examples and add a novel consistency regularization term to make better use of the misclassified examples. Specifically, we identified that the certified robustness of network can be significantly improved if the constraint of certified robustness on misclassified examples and correctly classified examples is consistent. Motivated by this discovery, we design a new defense regularization term called Misclassification Aware Adversarial Regularization (MAAR), which constrains the output probability distributions of all examples in the certified region of the misclassified example. Experimental results show that our proposed MAAR achieves the best certified robustness and comparable accuracy on CIFAR-10 and MNIST datasets in comparison with several state-of-the-art methods.
翻译:已经提出了一系列的防御方法,以提高神经网络在对抗性例子方面的稳健性,其中可证实的防御方法已证明对训练神经网络是有效的,对攻击者来说,这些可证明是稳健的。然而,大多数可证实的国防方法在培训过程中对所有例子一视同仁,忽视了正确分类(自然)和错误分类实例之间经核证的稳健性之间的不一致制约。在本文件中,我们探讨了错误分类实例造成的这种不一致,并增加了一个新的一致性规范术语,以便更好地使用错误分类的例子。具体地说,我们发现,如果对错误分类实例和正确分类实例的经证明的稳健性受到一致的限制,那么经证明的网络的稳健性是可以大大改进的。受这一发现驱使,我们设计了一个新的国防正规化术语,称为 " 认知错误分类 " (MAAR),它制约了认证错误分类实例所在区域所有实例的输出概率分布。实验结果表明,我们提议的MAAR在CAR-10和MNIST数据集方面获得了最佳的经证明的稳健性和可比的准确性。