无需训练模型，用后验指标判断数据是否被用于训练
Bayesian Inference of Training Dataset Membership
- 基于预测误差、置信度等后验指标进行贝叶斯推断
- 在合成数据集上准确区分训练集与非训练集数据
- 方法可解释且能检测分布偏移，适合隐私分析场景
确定某数据集是否曾被用于机器学习模型的训练，可揭示隐私漏洞，这通常通过成员推理攻击（MIAs）解决。传统MIAs需访问模型内部或依赖计算量大的影子模型。本文提出一种高效、可解释且原理严谨的贝叶斯推理方法，仅通过分析训练后模型的预测误差、置信度（熵）、扰动幅度及数据统计特征等后验指标，即可计算成员归属的后验概率，无需额外模型训练。在合成数据集上的实验表明，该方法能有效区分成员与非成员数据。此外，该方法还可检测分布偏移，为现有技术提供一种实用且可解释的替代方案。
原文摘要 · Abstract (English)
Determining whether a dataset was part of a machine learning model's training data pool can reveal privacy vulnerabilities, a challenge often addressed through membership inference attacks (MIAs). Traditional MIAs typically require access to model internals or rely on computationally intensive shadow models. This paper proposes an efficient, interpretable and principled Bayesian inference method for membership inference. By analyzing post-hoc metrics such as prediction error, confidence (entropy), perturbation magnitude, and dataset statistics from a trained ML model, our approach computes posterior probabilities of membership without requiring extensive model training. Experimental results on synthetic datasets demonstrate the method's effectiveness in distinguishing member from non-member datasets. Beyond membership inference, this method can also detect distribution shifts, offering a practical and interpretable alternative to existing approaches.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。