PulseAugur
中
实时 07:49:52
English(EN) Projection-Based Federated Learning on Compact Submanifolds with Geometry-Independent Differential Privacy

新的联邦学习方法增强了复杂流形上的差分隐私

研究人员开发了 DP-RFedProj,一种用于联邦学习的新方法,可增强紧凑子流形上的差分隐私。该方法在固定的环境几何结构中对梯度进行私有化,允许客户端执行多个本地步骤并间歇性参与,同时保持明确的记录级 f-DP 保证。该方法确立了黎曼 SGD 和投影平均的收敛性,隐私成本取决于流形的内在维度。在联邦 PCA 和 EEG 运动想象分类上的实验表明,与欧几里得基线相比,该方法具有更高的效用,并且在相同的隐私预算下所需的噪声明显更少。 AI

影响 增强了在复杂数据结构上进行联邦学习的隐私保证,可能实现更安全的协作模型训练。

排序理由 该集群包含一篇学术论文,详细介绍了具有差分隐私的新型联邦学习方法。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv stat.ML 阅读 →

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

新的联邦学习方法增强了复杂流形上的差分隐私

本文如何被排名

Signal score
22 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该集群包含一篇学术论文,详细介绍了具有差分隐私的新型联邦学习方法。[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。

报道来源 [1]

  1. arXiv stat.ML TIER_1 English(EN) · Thibault Pautrel, Florent Bouchard, Guillaume Ginolhac, Ammar Mian, Aur\'elien Bellet ·

    基于投影的紧凑子流形上的联邦学习与几何无关的差分隐私

    arXiv:2610.33712v1 Announce Type: new Abstract: Federated learning trains a shared model across clients without exchanging their data, yet the updates they transmit still leak information about individual records. Enforcing differential privacy is harder still when the model para…