arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2101.01341cs.CRcs.LG
收藏

基于差分比较的实用盲成员推断攻击

Practical Blind Membership Inference Attack via Differential Comparisons

The Johns Hopkins University(约翰斯·霍普金斯大学);The Johns Hopkins University Applied Physics Laboratory(约翰斯·霍普金斯大学应用物理实验室);Duke University(杜克大学)

Bo Hui, Yuchen Yang, Haolin Yuan, Philippe Burlina, Neil Zhenqiang Gong, Yinzhi Cao

摘要:

成员推断(MI)攻击通过推断给定数据样本是否已被用于训练目标学习模型来影响用户隐私,深度神经网络。文献中有两种类型的MI攻击,即,这些有和没有阴影模型。前者的成功很大程度上取决于阴影模型的质量,即,影子和目标之间的可转移性;后者,只给出黑盒探测访问目标模型,不能作出有效的未知数的推断,与使用影子模型的MI攻击相比,由于标记有地面真实成员信息的合格样本的数量不足。在本文中,我们提出了一个MI攻击,称为BlindMI,探测目标模型,并通过一种新的方法,称为差分比较提取成员语义。其高级思想是BlindMI首先通过将现有样本转换为新样本来生成具有非成员的数据集,然后以迭代方式将样本从目标数据集差异地移动到生成的非成员集。如果样本的差分移动增加了设置距离,则BlindMI将该样本视为非成员,反之亦然。通过将其与最先进的MI攻击算法进行比较来评估BlindMI。我们的评估表明,在对手不知道目标模型的架构和目标数据集的地面真实标签的盲设置中,与某些数据集(如Purchase-50和Birds-200)的最新技术相比,BlindMI将F1分数提高了近20%。我们还表明,BlindMI可以击败最先进的防御。

英文摘要:

Membership inference (MI) attacks affect user privacy by inferring whether given data samples have been used to train a target learning model, e.g., a deep neural network. There are two types of MI attacks in the literature, i.e., these with and without shadow models. The success of the former heavily depends on the quality of the shadow model, i.e., the transferability between the shadow and the target; the latter, given only blackbox probing access to the target model, cannot make an effective inference of unknowns, compared with MI attacks using shadow models, due to the insufficient number of qualified samples labeled with ground truth membership information. In this paper, we propose an MI attack, called BlindMI, which probes the target model and extracts membership semantics via a novel approach, called differential comparison. The high-level idea is that BlindMI first generates a dataset with nonmembers via transforming existing samples into new samples, and then differentially moves samples from a target dataset to the generated, non-member set in an iterative manner. If the differential move of a sample increases the set distance, BlindMI considers the sample as non-member and vice versa. BlindMI was evaluated by comparing it with state-of-the-art MI attack algorithms. Our evaluation shows that BlindMI improves F1-score by nearly 20% when compared to state-of-the-art on some datasets, such as Purchase-50 and Birds-200, in the blind setting where the adversary does not know the target model's architecture and the target dataset's ground truth labels. We also show that BlindMI can defeat state-of-the-art defenses.

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。

↑