PulseAugur
实时 03:44:16
English(EN) Differential Privacy of Gaussian Process Posterior Sampling

新论文探讨高斯过程和机器学习报告中的差分隐私

两篇最新的arXiv论文探讨了机器学习中的差分隐私,重点关注高斯过程和报告机制。第一篇论文详细介绍了高斯过程后验采样固有的随机性如何提供差分隐私保证,其界限取决于正则化和后验方差。第二篇论文提倡使用非渐近高斯差分隐私(GDP)作为传达DP-SGD等算法隐私保证的更准确方法,并指出其能够以最小的误差捕获完整的隐私配置文件。 AI

影响 这些论文有助于加深对机器学习中隐私的理论理解,并可能影响未来人工智能系统的隐私保证的开发和传达方式。

排序理由 两篇在arXiv上发表的学术论文,讨论了机器学习背景下的差分隐私。

在 arXiv stat.ML 阅读 →

AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →

新论文探讨高斯过程和机器学习报告中的差分隐私

报道来源 [3]

  1. arXiv stat.ML TIER_1 English(EN) · Tomasz Maciazek ·

    高斯过程后验采样中的差分隐私

    arXiv:2606.17995v1 Announce Type: new Abstract: We study the privacy of releasing posterior sample paths from a Gaussian process (GP) when the entire training set including covariates and responses is private. Unlike standard differential-privacy (DP) mechanisms that add external…

  2. arXiv stat.ML TIER_1 English(EN) · Juan Felipe Gomez, Bogdan Kulynych, Georgios Kaissis, Flavio P. Calmon, Jamie Hayes, Borja Balle, Antti Honkela ·

    用于报告机器学习中差分隐私保证的高斯差分隐私

    arXiv:2503.10945v3 Announce Type: replace-cross Abstract: Current practices for reporting differential privacy (DP) guarantees for machine learning (ML) algorithms such as DP-SGD provide an incomplete and potentially misleading picture. For instance, if only a single $(\varepsilo…

  3. arXiv stat.ML TIER_1 English(EN) · Tomasz Maciazek ·

    高斯过程后验采样中的差分隐私

    We study the privacy of releasing posterior sample paths from a Gaussian process (GP) when the entire training set including covariates and responses is private. Unlike standard differential-privacy (DP) mechanisms that add external noise, posterior sampling is random by construc…