arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.10686cs.ITcs.CRcs.LGeess.SPmath.IT
收藏

论不确定性与校准对似然比成员推理攻击的影响

On the Impact of Uncertainty and Calibration on Likelihood-Ratio Membership Inference Attacks

Beijing Key Laboratory of Network System Architecture and Convergence, School of Information and Communication Engineering, Beijing University of Posts and Telecommunications(北京网络系统架构与融合重点实验室,信息与通信工程学院,北京邮电大学);King’s Communications, Learning & Information Processing (KCLIP) lab, Department of Engineering, King’s College London(国王学院通讯、学习与信息处理(KCLIP)实验室,工程学院,伦敦国王学院)

Meiyi Zhu, Caili Guo, Chunyan Feng, Osvaldo Simeone

AI总结:

本文在信息论框架下分析LiRA,揭示任意不确定性、认知不确定性与模型校准对成员推理攻击优势的影响,并通过三种披露设置验证解析界。

摘要:

在成员推理攻击(MIA)中,攻击者利用典型机器学习模型表现出的过度自信,来判断某个特定数据点是否曾被用于训练目标模型。本文在一个信息论框架内分析似然比攻击(LiRA)的性能,该框架能够研究真实数据生成过程中的任意不确定性、有限训练数据集导致的认知不确定性,以及目标模型校准水平所产生的影响。我们比较了三种不同设置,在这些设置中,攻击者从目标模型获得的反馈信息量依次降低:置信度向量(CV)披露,即输出概率向量被公开;真实标签置信度(TLC)披露,即模型仅提供分配给真实标签的概率;以及决策集(DS)披露,即如同保形预测那样生成一个自适应预测集。我们推导了MIA攻击者优势的界,旨在深入理解不确定性和校准对MIA有效性的影响。仿真结果表明,所推导的解析界能够很好地预测MIA的有效性。

英文摘要:

In a membership inference attack (MIA), an attacker exploits the overconfidence exhibited by typical machine learning models to determine whether a specific data point was used to train a target model. In this paper, we analyze the performance of the likelihood ratio attack (LiRA) within an information-theoretical framework that allows the investigation of the impact of the aleatoric uncertainty in the true data generation process, of the epistemic uncertainty caused by a limited training data set, and of the calibration level of the target model. We compare three different settings, in which the attacker receives decreasingly informative feedback from the target model: confidence vector (CV) disclosure, in which the output probability vector is released; true label confidence (TLC) disclosure, in which only the probability assigned to the true label is made available by the model; and decision set (DS) disclosure, in which an adaptive prediction set is produced as in conformal prediction. We derive bounds on the advantage of an MIA adversary with the aim of offering insights into the impact of uncertainty and calibration on the effectiveness of MIAs. Simulation results demonstrate that the derived analytical bounds predict well the effectiveness of MIAs.

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。

↑