arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2501.08236cs.LG
收藏

面向机器学习的隐私保护模型与预处理验证

Privacy-Preserving Model and Preprocessing Verification for Machine Learning

Case Western Reserve University(凯斯西储大学);IBM Research(IBM研究院);UThealth;Rutgers University(罗格斯大学)

Wenbiao Li, Anisa Halimi, Xiaoqian Jiang, Jaideep Vaidya, Erman Ayday

AI总结:

本研究提出一种结合LDP与LIME、SHAP模型解释的隐私保护验证框架,可完成预处理相关的二分类与多分类验证任务,在三个真实数据集上验证了其检测错误的有效性与隐私保障能力。

摘要:

本文提出了一种用于机器学习模型隐私保护验证的框架,聚焦于在敏感数据上训练的模型。该框架将本地差分隐私(Local Differential Privacy, LDP)与来自LIME和SHAP的模型解释相结合,能够在不损害个体隐私的前提下实现鲁棒验证。它解决两项核心任务:二分类任务,即通过应用恰当的预处理步骤验证目标模型是否得到正确训练;多分类任务,用于识别具体的预处理错误。在Diabetes、Adult和Student Record三个真实世界数据集上的评估表明,基于机器学习的方法在二分类任务中效果尤为突出,而基于阈值的方法在多分类任务中表现相当。结果显示,尽管验证准确率随数据集和噪声水平有所不同,但该框架能够有效检测预处理错误,提供强隐私保障,在保护敏感数据方面具备实际应用价值。

英文摘要:

This paper presents a framework for privacy-preserving verification of machine learning models, focusing on models trained on sensitive data. Integrating Local Differential Privacy (LDP) with model explanations from LIME and SHAP, our framework enables robust verification without compromising individual privacy. It addresses two key tasks: binary classification, to verify if a target model was trained correctly by applying the appropriate preprocessing steps, and multi-class classification, to identify specific preprocessing errors. Evaluations on three real-world datasets-Diabetes, Adult, and Student Record-demonstrate that while the ML-based approach is particularly effective in binary tasks, the threshold-based method performs comparably in multi-class tasks. Results indicate that although verification accuracy varies across datasets and noise levels, the framework provides effective detection of preprocessing errors, strong privacy guarantees, and practical applicability for safeguarding sensitive data.

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。

↑