arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1802.09707stat.MLcs.CRcs.LG
收藏

理解和增强对抗性示例的可移植性

Understanding and Enhancing the Transferability of Adversarial Examples

Peking University(北京大学);Beijing Institute of Big Data (BIBDR)(北京大数据研究院);Princeton University(普林斯顿大学)

Lei Wu, Zhanxing Zhu, Cheng Tai, Weinan E

摘要:

众所周知,最先进的深度神经网络容易受到对抗性示例的攻击,这些示例是通过对原始输入施加微小但恶意的扰动而形成的。此外,扰动可以\textit{transfer across models}:为特定模型生成的对抗性示例通常会误导其他看不见的模型。因此,对手可以利用它来攻击部署的系统,而无需任何查询,这严重阻碍了深度学习的应用,特别是在安全性至关重要的领域。在这项工作中,我们系统地研究了两类可能影响对抗性示例可转移性的因素。一个是关于模型特定的因素,包括网络架构,模型容量和测试精度。另一个是构造对抗样本的损失函数的局部光滑性。在此基础上,提出了一种简单而有效的增强可移植性的策略。我们称之为方差缩减攻击,因为它利用方差缩减梯度来生成对抗性示例。在CIFAR-10和ImageNet数据集上进行的各种实验证实了其有效性。

英文摘要:

State-of-the-art deep neural networks are known to be vulnerable to adversarial examples, formed by applying small but malicious perturbations to the original inputs. Moreover, the perturbations can \textit{transfer across models}: adversarial examples generated for a specific model will often mislead other unseen models. Consequently the adversary can leverage it to attack deployed systems without any query, which severely hinder the application of deep learning, especially in the areas where security is crucial. In this work, we systematically study how two classes of factors that might influence the transferability of adversarial examples. One is about model-specific factors, including network architecture, model capacity and test accuracy. The other is the local smoothness of loss function for constructing adversarial examples. Based on these understanding, a simple but effective strategy is proposed to enhance transferability. We call it variance-reduced attack, since it utilizes the variance-reduced gradient to generate adversarial example. The effectiveness is confirmed by a variety of experiments on both CIFAR-10 and ImageNet datasets.

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。

↑