arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2111.01080cs.LGcs.CR
收藏

ZeBRA:基于零数据的重复位翻转攻击精确破坏神经网络

ZeBRA: Precisely Destroying Neural Networks with Zero-Data Based Repeated Bit Flip Attack

Hongik University(弘益大学)

Dahoon Park, Kon-Woo Kwon, Sunghoon Im, Jaeha Kung

摘要:

在本文中,我们提出了基于零数据的重复位翻转攻击(ZeBRA),通过合成自己的攻击数据集来精确破坏深度神经网络(DNN)。许多对抗性权重攻击的现有工作不仅需要权重参数,而且需要训练或测试数据集来搜索要攻击的脆弱比特。我们建议通过利用受害者DNN模型中的批量归一化层的统计数据来合成攻击数据集,即提取的目标数据。配备了提取的目标数据,我们的ZeBRA算法可以搜索模型中的脆弱位,而无需访问训练或测试数据集。因此,我们的方法使得对抗性权重攻击对DNN的安全性更加致命。我们的实验结果表明,与以前的攻击方法相比,平均需要2.0倍(CIFAR-10)和1.6倍(ImageNet)的位翻转次数来破坏DNN。我们的代码可以在https://github上找到。com/pdh930105/ZeBRA。

英文摘要:

In this paper, we present Zero-data Based Repeated bit flip Attack (ZeBRA) that precisely destroys deep neural networks (DNNs) by synthesizing its own attack datasets. Many prior works on adversarial weight attack require not only the weight parameters, but also the training or test dataset in searching vulnerable bits to be attacked. We propose to synthesize the attack dataset, named distilled target data, by utilizing the statistics of batch normalization layers in the victim DNN model. Equipped with the distilled target data, our ZeBRA algorithm can search vulnerable bits in the model without accessing training or test dataset. Thus, our approach makes the adversarial weight attack more fatal to the security of DNNs. Our experimental results show that 2.0x (CIFAR-10) and 1.6x (ImageNet) less number of bit flips are required on average to destroy DNNs compared to the previous attack method. Our code is available at https://github. com/pdh930105/ZeBRA.

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。

↑