超越对抗性扰动：具有合法语义的流形辅助对抗样本
Transcending Adversarial Perturbations: Manifold-Aided Adversarial Examples with Legitimate Semantics
AI总结：
本文提出监督式语义变换生成模型，构建无约束对抗流形生成合法语义对抗样本，在MNIST和工业缺陷数据集上实现更优视觉质量、攻击可迁移性和可解释性。
摘要：
深度神经网络极易受到由恶意微小扰动操控的对抗样本的攻击。尽管大多数传统对抗攻击通过最小化对抗样本与原始图像之间的几何距离来确保视觉上的不可感知性，但这些几何距离约束导致了攻击可迁移性受限、视觉质量较差以及人类难以感知的可解释性。在本文中，我们提出了一种监督式语义变换生成模型，用于生成具有真实且合法语义的对抗样本，其中首次构建了一个包含连续语义变化的无约束对抗流形，以实现从非对抗样本到对抗样本的合法过渡。在MNIST和工业缺陷数据集上的全面实验表明，我们的对抗样本不仅展现出更好的视觉质量，还实现了更优的攻击可迁移性以及对模型漏洞更有效的解释，表明其作为通用对抗样本的巨大潜力。代码和预训练模型可在 https://github.com/shuaili1027/MAELS.git 获取。
英文摘要：
Deep neural networks were significantly vulnerable to adversarial examples manipulated by malicious tiny perturbations. Although most conventional adversarial attacks ensured the visual imperceptibility between adversarial examples and corresponding raw images by minimizing their geometric distance, these constraints on geometric distance led to limited attack transferability, inferior visual quality, and human-imperceptible interpretability. In this paper, we proposed a supervised semantic-transformation generative model to generate adversarial examples with real and legitimate semantics, wherein an unrestricted adversarial manifold containing continuous semantic variations was constructed for the first time to realize a legitimate transition from non-adversarial examples to adversarial ones. Comprehensive experiments on MNIST and industrial defect datasets showed that our adversarial examples not only exhibited better visual quality but also achieved superior attack transferability and more effective explanations for model vulnerabilities, indicating their great potential as generic adversarial examples. The code and pre-trained models were available at https://github.com/shuaili1027/MAELS.git.
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。
