...
首页> 外文期刊>JMLR: Workshop and Conference Proceedings >A Fast Spectral Algorithm for Mean Estimation with Sub-Gaussian Rates
【24h】

A Fast Spectral Algorithm for Mean Estimation with Sub-Gaussian Rates

机译:具有亚高斯速率的平均估计快速谱算法

获取原文
           

摘要

We study the algorithmic problem of estimating the mean of a heavy-tailed random vector in R^d, given n i.i.d. samples. The goal is to design an efficient estimator that attains the optimal sub-gaussian error bound, only assuming that the random vector has bounded mean and covariance. Polynomial-time solutions to this problem are known but have high runtime due to their use of semi-definite programming (SDP). Moreover, conceptually, it remains open whether convex relaxation is truly necessary for this problem. In this work, we show that it is possible to go beyond SDP and achieve better computational efficiency. In particular, we provide a spectral algorithm that achieves the optimal statistical performance and runs in time O?( n^2 d ), improving upon the previous fastest runtime O( n^{3.5}+ n^2 d ) by Cherapanamjeri et.al. (COLT ’19). Our algorithm is spectral in that it only requires (approximate) eigenvector computations, which can be implemented very efficiently by, for example, power iteration or the Lanczos method. At the core of our algorithm is a novel connection between the furthest hyperplane problem introduced by Karnin et. al. (COLT ’12) and a structural lemma on heavy-tailed distributions by Lugosi and Mendelson (Ann. Stat. ’19). This allows us to iteratively reduce the estimation error at a geometric rate using only the information derived from the top singular vector of the data matrix, leading to a significantly faster running time.
机译:我们研究了估计R ^ D中的重尾随机载体的平均值的算法问题,给出了n i.i.d.样品。目标是设计一个有效的估计器,该估计器仅达到最佳的子高斯误差绑定,仅假设随机向量具有界定均值和协方差。由于它们使用半定编程(SDP),该问题的多项式时间解决方案是已知的,但具有高运行时。此外,概念上,它仍然打开凸面放松是真的需要这个问题。在这项工作中,我们表明可以超越SDP并实现更好的计算效率。特别是,我们提供一种频谱算法,实现了最佳统计性能,并在时间o?(n ^ 2 d),通过cherapanamjeri et提高了前一个最快的运行时o(n ^ {3.5} + n ^ 2 d)。 al。 (Colt'19)。我们的算法是频谱,因为它仅需要(近似)特征向量计算,其可以通过例如功率迭代或LanczoS方法非常有效地实现。在我们的算法的核心,是Karnin等引入的最远超平面问题之间的新连接。 al。 (Colt'12)和卢比索和孟德尔森的重型分布上的结构引理(ANN。统计数据。'19)。这允许我们仅使用从数据矩阵的顶部奇异向量导出的信息来迭代地以几何速率降低估计误差,从而导致更快的运行时间。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号