对抗性划痕：针对 CNN 分类器的可部署攻击
Adversarial Scratches: Deployable Attacks to CNN Classifiers
AI总结：
提出一种新型可部署黑盒攻击方法 Adversarial Scratches，通过 Bézier 曲线生成图像划痕扰动，在降低查询次数和修改像素数的同时实现更高的欺骗率。
摘要：
越来越多的研究表明，深度神经网络易受对抗样本攻击。这些攻击通过对模型输入施加微小扰动，导致模型做出错误预测。然而，现有文献大多关注应用于数字图像的视觉不可察觉扰动，这类扰动通常在设计上难以部署到物理目标。我们提出了一种新型 L0 黑盒攻击方法——Adversarial Scratches（对抗性划痕），其以图像中的划痕形式呈现，可部署性显著优于其他最先进攻击方法。该方法利用 Bézier 曲线降低搜索空间维度，并可将攻击约束在特定位置。我们在多种场景下测试了该方法，包括公开 API 和交通标志图像。实验结果表明，与现有可部署方法相比，本攻击通常能达到更高的欺骗率，且所需查询次数显著更少，修改像素数极少。
英文摘要：
A growing body of work has shown that deep neural networks are susceptible to adversarial examples. These take the form of small perturbations applied to the model's input which lead to incorrect predictions. Unfortunately, most literature focuses on visually imperceivable perturbations to be applied to digital images that often are, by design, impossible to be deployed to physical targets. We present Adversarial Scratches: a novel L0 black-box attack, which takes the form of scratches in images, and which possesses much greater deployability than other state-of-the-art attacks. Adversarial Scratches leverage Bézier Curves to reduce the dimension of the search space and possibly constrain the attack to a specific location. We test Adversarial Scratches in several scenarios, including a publicly available API and images of traffic signs. Results show that, often, our attack achieves higher fooling rate than other deployable state-of-the-art methods, while requiring significantly fewer queries and modifying very few pixels.
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。
