arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2005.02313cs.CVcs.CRcs.LGstat.ML
收藏

针对位置优化对抗补丁的对抗训练

Adversarial Training against Location-Optimized Adversarial Patches

Max Planck Institute for Informatics(马克斯·普朗克信息学研究所)

Sukrut Rao, David Stutz, Bernt Schiele

AI总结:

本研究提出了一种优化对抗补丁位置的方法,并对其应用对抗训练,在CIFAR10和GTSRB上显著提升鲁棒性且不降低准确率。

摘要:

深度神经网络已被证明容易受到对抗样本的影响——这些是精心构造的微小、不可察觉的变化,旨在导致原本高度准确的图像分类器产生误分类。作为一种实用的替代方案,最近的工作提出了所谓的对抗补丁:图像中清晰可见但经过对抗性精心设计的矩形补丁。这些补丁可以轻松打印并应用于物理世界。虽然针对不可察觉对抗样本的防御已被广泛研究,但对对抗补丁的鲁棒性仍知之甚少。在这项工作中,我们首先设计了一种实用方法来获取对抗补丁,同时主动优化其在图像中的位置。然后,我们将对抗训练应用于这些位置优化的对抗补丁,并在CIFAR10和GTSRB上展示了显著提升的鲁棒性。此外,与针对不可察觉对抗样本的对抗训练不同,我们的对抗补丁训练不会降低准确率。

英文摘要:

Deep neural networks have been shown to be susceptible to adversarial examples -- small, imperceptible changes constructed to cause mis-classification in otherwise highly accurate image classifiers. As a practical alternative, recent work proposed so-called adversarial patches: clearly visible, but adversarially crafted rectangular patches in images. These patches can easily be printed and applied in the physical world. While defenses against imperceptible adversarial examples have been studied extensively, robustness against adversarial patches is poorly understood. In this work, we first devise a practical approach to obtain adversarial patches while actively optimizing their location within the image. Then, we apply adversarial training on these location-optimized adversarial patches and demonstrate significantly improved robustness on CIFAR10 and GTSRB. Additionally, in contrast to adversarial training on imperceptible adversarial examples, our adversarial patch training does not reduce accuracy.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑