新型优化器DP-IVON-Gradsq增强了贝叶斯深度学习中的差分隐私

作者 PulseAugur 编辑部 · [1 个来源] · 2026-07-28 04:00

研究人员开发了DP-IVON-Gradsq，这是一种旨在增强贝叶斯深度学习中差分隐私的新型优化器。该方法旨在减轻隐私噪声与贝叶斯后验采样固有的随机性之间的干扰。通过使用噪声校正的平方梯度估计器，DP-IVON-Gradsq在保持与Adam相似的计算效率的同时，增强了隐私保证。在CIFAR-10数据集上的评估表明，在隐私约束不那么严格的情况下，DP-IVON-Gradsq的性能与DP-SGD和DP-Adam相当。 AI

影响引入了一种新型优化器，可以改进贝叶斯深度学习模型的隐私保护训练。

排序理由该集群包含一篇研究论文，详细介绍了贝叶斯深度学习中差分隐私的新方法。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.LG 阅读 →

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; 该集群包含一篇研究论文，详细介绍了贝叶斯深度学习中差分隐私的新方法。[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: 72 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) · Nour Jamoussi, Ikram Dridi, Giuseppe Serra, Marios Kountouris · 2026-07-28 04:00

DP-IVON-Gradsq：差分隐私平方梯度改进变分在线牛顿法

arXiv:2607.23649v1 Announce Type: new Abstract: Differential privacy provides formal privacy guarantees for training neural networks on sensitive data, while Bayesian deep learning offers a principled framework for uncertainty-aware prediction. Combining these two objectives rema…

本文如何被排名

报道来源 [1]

DP-IVON-Gradsq：差分隐私平方梯度改进变分在线牛顿法

相关实体

相关话题