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 2201.13164cs.CV
收藏

针对深度神经网络的不可感知且多通道的后门攻击

Imperceptible and Multi-channel Backdoor Attack against Deep Neural Networks

College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics(南京航空航天大学计算机科学与技术学院);College of Electronic and Information Engineering, Nanjing University of Aeronautics and Astronautics(南京航空航天大学电子信息工程学院)

Mingfu Xue, Shifeng Ni, Yinghao Wu, Yushu Zhang, Jian Wang, Weiqiang Liu

AI总结:

本文利用DCT隐写术提出一种不可感知且多通道的后门攻击方法,实现了N-to-N和N-to-One两种变体,在保持模型准确率的同时具有高攻击成功率且对Neural Cleanse防御具有鲁棒性。

摘要:

最近的研究表明,深度神经网络(DNN)模型容易受到后门攻击。当包含后门触发器的图像到达时,被植入后门的DNN模型会表现出恶意行为。迄今为止,现有的后门攻击均为单触发器和单目标攻击,且大多数现有后门攻击的触发器非常明显,因此很容易被检测到或注意到。在本文中,我们通过利用离散余弦变换(DCT)隐写术,提出了一种针对深度神经网络的不可感知且多通道的后门攻击。基于所提出的后门攻击方法,我们实现了两种后门攻击变体,即N-to-N后门攻击和N-to-One后门攻击。具体而言,对于彩色图像,我们利用DCT隐写术在图像的不同通道上构建触发器。因此,该触发器具有隐蔽性和自然性。基于所提出的方法,我们实现了多目标、多触发器的后门攻击。实验结果表明,N-to-N后门攻击在CIFAR-10数据集上的平均攻击成功率为93.95%,在TinyImageNet数据集上为91.55%。N-to-One攻击在CIFAR-10和TinyImageNet数据集上的平均攻击成功率分别为90.22%和89.53%。同时,所提出的后门攻击不会影响DNN模型的分类准确率。此外,所提出的攻击被证明对最先进的后门防御(Neural Cleanse)具有鲁棒性。

英文摘要:

Recent researches demonstrate that Deep Neural Networks (DNN) models are vulnerable to backdoor attacks. The backdoored DNN model will behave maliciously when images containing backdoor triggers arrive. To date, existing backdoor attacks are single-trigger and single-target attacks, and the triggers of most existing backdoor attacks are obvious thus are easy to be detected or noticed. In this paper, we propose a novel imperceptible and multi-channel backdoor attack against Deep Neural Networks by exploiting Discrete Cosine Transform (DCT) steganography. Based on the proposed backdoor attack method, we implement two variants of backdoor attacks, i.e., N-to-N backdoor attack and N-to-One backdoor attack. Specifically, for a colored image, we utilize DCT steganography to construct the trigger on different channels of the image. As a result, the trigger is stealthy and natural. Based on the proposed method, we implement multi-target and multi-trigger backdoor attacks. Experimental results demonstrate that the average attack success rate of the N-to-N backdoor attack is 93.95% on CIFAR-10 dataset and 91.55% on TinyImageNet dataset, respectively. The average attack success rate of N-to-One attack is 90.22% and 89.53% on CIFAR-10 and TinyImageNet datasets, respectively. Meanwhile, the proposed backdoor attack does not affect the classification accuracy of the DNN model. Moreover, the proposed attack is demonstrated to be robust to the state-of-the-art backdoor defense (Neural Cleanse).

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。

↑