arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2104.09425cs.LGcs.CRcs.CV
收藏

鲁棒学习满足生成模型:代理分布可以提高对抗鲁棒性吗?

Robust Learning Meets Generative Models: Can Proxy Distributions Improve Adversarial Robustness?

Princeton University(普林斯顿大学);Caltech(加州理工学院);Purdue University(普渡大学)

Vikash Sehwag, Saeed Mahloujifar, Tinashe Handina, Sihui Dai, Chong Xiang, Mung Chiang, Prateek Mittal

摘要:

虽然额外的训练数据提高了深度神经网络对对抗性示例的鲁棒性,但它带来了管理大量特定真实世界样本的挑战。我们通过使用来自高级生成模型学习的代理分布的额外数据来规避这一挑战。我们首先寻求正式理解的转移鲁棒性从代理分布训练的分类器的真实数据分布。我们证明了两个分布的分类器的鲁棒性之间的差异是上界的条件Wasserstein距离。接下来,我们使用代理分布来显著提高对抗训练在五个不同数据集上的性能。例如,我们在$\ell_{\infty}$和$\ell_2$威胁模型中将鲁棒准确性提高了7.5%和6.7%,超过了在CIFAR-10数据集上不使用代理分布的基线。我们还将CIFAR-10数据集的认证稳健准确率提高了7.6%。我们进一步证明了不同的生成模型在鲁棒训练中带来了不同的性能改善。我们提出了一个强大的判别方法来表征个体生成模型的影响,并进一步深入了解为什么当前基于扩散的生成模型是代理分布的更好选择,而不是生成对抗网络。

英文摘要:

While additional training data improves the robustness of deep neural networks against adversarial examples, it presents the challenge of curating a large number of specific real-world samples. We circumvent this challenge by using additional data from proxy distributions learned by advanced generative models. We first seek to formally understand the transfer of robustness from classifiers trained on proxy distributions to the real data distribution. We prove that the difference between the robustness of a classifier on the two distributions is upper bounded by the conditional Wasserstein distance between them. Next we use proxy distributions to significantly improve the performance of adversarial training on five different datasets. For example, we improve robust accuracy by up to 7.5% and 6.7% in $\ell_{\infty}$ and $\ell_2$ threat model over baselines that are not using proxy distributions on the CIFAR-10 dataset. We also improve certified robust accuracy by 7.6% on the CIFAR-10 dataset. We further demonstrate that different generative models bring a disparate improvement in the performance in robust training. We propose a robust discrimination approach to characterize the impact of individual generative models and further provide a deeper understanding of why current state-of-the-art in diffusion-based generative models are a better choice for proxy distribution than generative adversarial networks.

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。

↑