English(EN) The Sample Complexity of Membership Inference and Privacy Auditing

新研究表明成员推断攻击可能比之前认为需要更多数据

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-09 04:00

一篇新研究论文探讨了成员推断攻击所需的样本复杂度，该攻击旨在确定个体数据是否被用于训练机器学习模型。该研究聚焦于高斯均值估计，表明攻击者可能需要比训练算法本身用于训练模型数量多得多的参考样本。这一发现挑战了当前实际攻击，这些攻击仅限于使用与训练数据大小成比例的样本数量，并可能低估隐私泄露的可能性。 AI

影响这项研究突显了当前成员推断攻击方法可能低估隐私风险，表明需要更强大的防御措施。

排序理由该集群包含一篇详细介绍成员推断攻击理论发现的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

Mahdi Haghifam

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群包含一篇详细介绍成员推断攻击理论发现的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, safety; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 15 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.LG TIER_1 English(EN) · Mahdi Haghifam, Adam Smith, Jonathan Ullman · 2026-10-09 04:00

成员推断和隐私审计的样本复杂度

arXiv:2508.19458v2 Announce Type: replace Abstract: A membership-inference attack gets the output of a learning algorithm, and a target individual, and tries to determine whether this individual is a member of the training data or an independent sample from the same distribution.…

本文如何被排名

报道来源 [1]

成员推断和隐私审计的样本复杂度

相关话题