arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.05748cs.AI
收藏

LeakBoost:基于感知损失的成员推断攻击

LeakBoost: Perceptual-Loss-Based Membership Inference Attack

Ben-Gurion University(本· Gurion大学)

Amit Kravchik Taub, Fred M. Grabovski, Guy Amit, Yisroel Mirsky

AI总结:

LeakBoost通过感知损失主动探测模型内部表示,提升成员推断攻击的准确性和效率,显著提高AUC和TPR,适用于多种神经网络架构。

摘要:

成员推断攻击(MIAs)旨在确定样本是否是模型训练集的一部分,对现代机器学习系统构成严重的隐私风险。现有MIAs主要依赖静态指标,如损失或置信度,并未充分利用在主动探测时模型的动态行为。我们提出LeakBoost,一种基于感知损失的探测框架,通过主动探测模型的内部表示来暴露隐藏的成员信号。给定一个候选输入,LeakBoost通过优化感知(激活空间)目标合成一个探测图像,放大成员和非成员之间的表示差异。该图像随后由现成的成员检测器分析,而无需修改检测器本身。当与现有成员推断方法结合时,LeakBoost在多个图像分类数据集和多样化的神经网络架构上实现了显著改进,在低误报率下取得优异效果。特别是,它将AUC从接近偶然水平(0.53-0.62)提升到0.81-0.88,并在1%误报率下将TPR提高一个数量级,相比强基线攻击。详细的敏感性分析揭示,更深的层和短的、低学习率的优化产生最强的泄露,且改进集中在基于梯度的检测器中。LeakBoost因此提供了一种模块化且计算高效的白盒环境下评估隐私风险的方法,推动了动态成员推断研究的发展。

英文摘要:

Membership inference attacks (MIAs) aim to determine whether a sample was part of a model's training set, posing serious privacy risks for modern machine-learning systems. Existing MIAs primarily rely on static indicators, such as loss or confidence, and do not fully leverage the dynamic behavior of models when actively probed. We propose LeakBoost, a perceptual-loss-based interrogation framework that actively probes a model's internal representations to expose hidden membership signals. Given a candidate input, LeakBoost synthesizes an interrogation image by optimizing a perceptual (activation-space) objective, amplifying representational differences between members and non-members. This image is then analyzed by an off-the-shelf membership detector, without modifying the detector itself. When combined with existing membership inference methods, LeakBoost achieves substantial improvements at low false-positive rates across multiple image classification datasets and diverse neural network architectures. In particular, it raises AUC from near-chance levels (0.53-0.62) to 0.81-0.88, and increases TPR at 1 percent FPR by over an order of magnitude compared to strong baseline attacks. A detailed sensitivity analysis reveals that deeper layers and short, low-learning-rate optimization produce the strongest leakage, and that improvements concentrate in gradient-based detectors. LeakBoost thus offers a modular and computationally efficient way to assess privacy risks in white-box settings, advancing the study of dynamic membership inference.

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。

↑