arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.12385cs.DScs.LG
收藏

更精确的Chebyshev矩匹配界限及其应用

Sharper Bounds for Chebyshev Moment Matching, with Applications

UMass Amherst(马萨诸塞大学阿姆赫斯特分校);New York University(纽约大学)

Cameron Musco, Christopher Musco, Lucas Rosenblatt, Apoorv Vikram Singh

AI总结:

本文研究了在存在噪声测量的情况下,通过Chebyshev多项式矩来近似恢复概率分布的问题。通过利用Lipschitz函数Chebyshev展开系数的全局衰减界,作者证明了在比之前已知的更多的噪声情况下,可以在Wasserstein距离中实现精确的恢复。该结果立即应用于多个领域:1)提供了一个简单的“线性查询”算法,用于构造具有Wasserstein-1误差为~O(1/n)的差分隐私合成数据分布;2)给出了一个~O(n²/ε)时间的算法,用于估计对称矩阵的谱密度,误差在Wasserstein距离内为ε;3)改进了Vinayak等人在ICML 2019上对“学习参数群体”统计问题最大似然估计器的分析,扩展了可以获得样本最优结果的参数范围。此外,作者还扩展了该界到d>1维分布的估计。

摘要:

我们研究了在存在噪声测量的情况下,通过Chebyshev多项式矩近似恢复概率分布的问题。这个问题在算法、统计和机器学习中广泛出现。通过利用任何Lipschitz函数Chebyshev展开系数的全局衰减界,我们改进了先前的工作,证明在比之前已知的更多的噪声情况下,可以在Wasserstein距离中实现精确的恢复。我们的结果立即导致了多个应用:1)我们提供了一个简单的“线性查询”算法,用于构造具有Wasserstein-1误差~O(1/n)的差分隐私合成数据分布,该结果在对数因子范围内是最佳的,并与Boedihardjo、Strohmer和Vershynin [Probab. Theory. Rel., 2024] 的结果相匹配,该结果使用了更复杂的“超正则随机游走”方法。2)我们给出了一个~O(n²/ε)时间的算法,用于估计n×n对称矩阵的谱密度,误差在Wasserstein距离内为ε。我们的结果加速了Chen等人[ICML 2021]和Braverman等人[STOC 2022]的先前方法。3)我们改进了Vinayak、Kong、Valiant和Kakade [ICML 2019] 对“学习参数群体”统计问题最大似然估计器的分析,扩展了可以获得样本最优结果的参数范围。除了这些主要结果外,我们还扩展了该界到d>1维分布的估计。我们希望这些界能更广泛地应用于涉及从噪声矩信息中恢复分布的问题。

英文摘要:

We study the problem of approximately recovering a probability distribution given noisy measurements of its Chebyshev polynomial moments. This problem arises broadly across algorithms, statistics, and machine learning. By leveraging a global decay bound on the coefficients in the Chebyshev expansion of any Lipschitz function, we sharpen prior work, proving that accurate recovery in the Wasserstein distance is possible with more noise than previously known. Our result immediately yields a number of applications: 1) We give a simple "linear query" algorithm for constructing a differentially private synthetic data distribution with Wasserstein-$1$ error $\tilde{O}(1/n)$ based on a dataset of $n$ points in $[-1,1]$. This bound is optimal up to log factors, and matches a recent result of Boedihardjo, Strohmer, and Vershynin [Probab. Theory. Rel., 2024], which uses a more complex "superregular random walk" method. 2) We give an $\tilde{O}(n^2/ε)$ time algorithm for the linear algebraic problem of estimating the spectral density of an $n\times n$ symmetric matrix up to $ε$ error in the Wasserstein distance. Our result accelerates prior methods from Chen et al. [ICML 2021] and Braverman et al. [STOC 2022]. 3) We tighten an analysis of Vinayak, Kong, Valiant, and Kakade [ICML 2019] on the maximum likelihood estimator for the statistical problem of "Learning Populations of Parameters'', extending the parameter regime in which sample optimal results can be obtained. Beyond these main results, we provide an extension of our bound to estimating distributions in $d > 1$ dimensions. We hope that these bounds will find applications more broadly to problems involving distribution recovery from noisy moment information.

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。

↑