标题： 使用条件生成模型理解图像分布的内在鲁棒性 显示英文标题

标题： Understanding the Intrinsic Robustness of Image Distributions using Conditional Generative Models

摘要： 从Gilmer等人(2018)开始，许多工作基于对底层输入概率空间的不同假设，证明了对抗样本的不可避免性。然而，这些结果是否适用于自然图像分布仍不清楚。在本工作中，我们假设底层数据分布由某种条件生成模型捕获，并证明了一类通用分类器的内在鲁棒性边界，解决了Fawzi等人(2018)中的一个开放问题。基于最先进的条件生成模型，我们研究了两种常见图像基准在$\ell_2$扰动下的内在鲁棒性，并展示了由我们的理论暗示的鲁棒性极限与当前最先进鲁棒模型实现的对抗鲁棒性之间的巨大差距。我们所有实验的代码可在https://github.com/xiaozhanguva/Intrinsic-Rob获取。 显示英文摘要

摘要： Starting with Gilmer et al. (2018), several works have demonstrated the inevitability of adversarial examples based on different assumptions about the underlying input probability space. It remains unclear, however, whether these results apply to natural image distributions. In this work, we assume the underlying data distribution is captured by some conditional generative model, and prove intrinsic robustness bounds for a general class of classifiers, which solves an open problem in Fawzi et al. (2018). Building upon the state-of-the-art conditional generative models, we study the intrinsic robustness of two common image benchmarks under $\ell_2$ perturbations, and show the existence of a large gap between the robustness limits implied by our theory and the adversarial robustness achieved by current state-of-the-art robust models. Code for all our experiments is available at https://github.com/xiaozhanguva/Intrinsic-Rob.