研究人员开发了PATE-TabTransGAN，这是一个用于生成符合正式差分隐私保证的合成表格数据的新颖框架。该方法结合了Teacher Ensembles的私有聚合（PATE）机制和基于Transformer的学生判别器，以在确保强大隐私的同时有效建模特征间的依赖关系。在四个基准数据集上的实验表明，与现有的最先进的差分隐私合成方法相比，PATE-TabTransGAN在AUROC和AUCPR方面取得了具有竞争力或更优的性能。 AI
影响 这项研究推动了隐私保护合成数据生成的最先进技术，可能有助于更安全地使用敏感表格数据集。
排序理由 该集群描述了一篇详细介绍差分隐私合成数据生成新方法的最新研究论文。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
arXiv:2605.26802v1 Announce Type: new Abstract: Generating high-fidelity synthetic tabular data under formal differential privacy guarantees remains an open challenge. Methods that provide strong theoretical protection typically sacrifice the modeling of inter-feature dependencie…
Generating high-fidelity synthetic tabular data under formal differential privacy guarantees remains an open challenge. Methods that provide strong theoretical protection typically sacrifice the modeling of inter-feature dependencies required for realistic synthesis, while archit…
Generating high-fidelity synthetic tabular data under formal differential privacy guarantees remains an open challenge. Methods that provide strong theoretical protection typically sacrifice the modeling of inter-feature dependencies required for realistic synthesis, while archit…