arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2609.02880cs.CRcs.CC

克服差分隐私线性查询中的随机性-效用权衡

Overcoming the Randomness-Utility Trade-off in Answering Differentially Private Linear Queries

  • Cornell University(康奈尔大学)
  • Google Research(谷歌研究院)

机构由 AI 辅助整理,请以论文原文为准。

Surendra Ghentiyala, Pritish Kamath, Ravi Kumar, Pasin Manurangsi

AI总结:

该研究针对差分隐私线性查询,提出了随机高效的‖·‖_K范数机制,其在ℓ_∞误差下优于现有算法且在ε≤1/d时最优,还给出了计算高效的版本。

AI中文摘要:

我们研究使用少量(期望)随机位实现差分隐私线性查询回答的问题。我们提出了Hardt和Talwar[HT10]的‖·‖_K范数机制的随机高效类似版本。对于ℓ_∞误差,我们的算法可回答d个线性查询,误差为O(d/ε),使用O(log d)个随机位,优于Canonne等人和Ghentiyala[CSV25, Ghe26]的算法;当ε≤1/d时,该结果是最优的。我们还提供了算法的计算高效版本,不过误差会有O(log d)的乘法级增加。

英文摘要:

We study the question of answering linear queries with differential privacy using few (expected) random bits. We provide a randomness-efficient analog of the $\| \cdot \|_K$-norm mechanism of Hardt and Talwar [HT10]. For the $\ell_\infty$-error, our algorithm can answer $d$ linear queries with $O(d / \varepsilon)$ error using $O(\log d)$ random bits, improving upon algorithms of Canonne et al. and Ghentiyala [CSV25, Ghe26]; this is optimal when $\varepsilon \le 1/d$. We also provide a computationally efficient version of our algorithm, albeit with an $O(\log d)$ multiplicative increase in the error.

↑