English(EN) Gradient Inversion through Linear and Affine Subspace Constraints

新的梯度反演攻击利用了联邦学习中的子空间约束

作者 PulseAugur 编辑部 · [1 个来源] · 2026-11-11 05:00

研究人员开发了新的联邦学习梯度反演攻击，利用了线性和仿射子空间约束。这些方法提高了从交换梯度中重建敏感客户端数据的准确性。研究还发现，适度高估批次大小可以提高重建质量，即使在不知道真实批次大小的情况下也是如此，并考察了差分隐私机制对攻击有效性的影响。 AI

影响增强了对联邦学习中隐私风险以及数据重建防御措施的理解。

排序理由学术论文，详细介绍了联邦学习中梯度反演攻击的新方法。[lever_c_demoted from research: ic=1 ai=1.0]

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: 36 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 学术论文，详细介绍了联邦学习中梯度反演攻击的新方法。[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: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

arXiv cs.LG TIER_1 English(EN) · Konstantinos Bouros, Athanasios P. Liavas · 2026-11-11 05:00

通过线性和仿射子空间约束实现梯度反演

arXiv:2611.13976v1 Announce Type: cross Abstract: Federated learning enables multiple clients to collaboratively train a machine-learning model without directly sharing their local datasets. However, as prior works have shown, the gradients exchanged during training may reveal se…