arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2306.04444cs.LGcs.CRstat.ML
收藏

通过随机投影实现快速最优的局部隐私均值估计

Fast Optimal Locally Private Mean Estimation via Random Projections

Apple Inc.(苹果公司);UC Berkeley(加州大学伯克利分校);Northeastern(东北大学)

Hilal Asi, Vitaly Feldman, Jelani Nelson, Huy L. Nguyen, Kunal Talwar

AI总结:

针对高维局部隐私均值估计中现有算法误差次优或复杂度高的问题,提出 ProjUnit 框架,通过随机投影至低维子空间并运行最优算法,在保证最优误差的同时显著降低通信与计算成本。

摘要:

我们研究欧氏球中高维向量的局部隐私均值估计问题。针对该问题的现有算法要么产生次优误差,要么具有较高的通信和/或运行时间复杂度。我们提出了一种新的算法框架 ProjUnit,用于隐私均值估计,其生成的算法计算高效、通信复杂度低,且产生的误差在 $1+o(1)$ 因子内是最优的。我们的框架极其简单:每个随机化器将其输入投影到一个随机的低维子空间,对结果进行归一化,然后在低维空间中运行诸如 PrivUnitG 之类的最优算法。此外,我们表明,通过适当地关联各设备间的随机投影矩阵,可以实现快速的服务器运行时间。我们根据随机投影的性质在数学上分析了算法的误差,并研究了两种实例化。最后,我们在隐私均值估计和隐私联邦学习上的实验表明,我们的算法在经验上获得了与最优算法几乎相同的效用,同时具有显著更低的通信和计算成本。

英文摘要:

We study the problem of locally private mean estimation of high-dimensional vectors in the Euclidean ball. Existing algorithms for this problem either incur sub-optimal error or have high communication and/or run-time complexity. We propose a new algorithmic framework, ProjUnit, for private mean estimation that yields algorithms that are computationally efficient, have low communication complexity, and incur optimal error up to a $1+o(1)$-factor. Our framework is deceptively simple: each randomizer projects its input to a random low-dimensional subspace, normalizes the result, and then runs an optimal algorithm such as PrivUnitG in the lower-dimensional space. In addition, we show that, by appropriately correlating the random projection matrices across devices, we can achieve fast server run-time. We mathematically analyze the error of the algorithm in terms of properties of the random projections, and study two instantiations. Lastly, our experiments for private mean estimation and private federated learning demonstrate that our algorithms empirically obtain nearly the same utility as optimal ones while having significantly lower communication and computational cost.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑