arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2307.08706cs.CGcs.DScs.LG
收藏

用于分位数回归的高效强多项式算法

Efficient Strongly Polynomial Algorithms for Quantile Regression

Google(谷歌);University of Illinois Chicago(伊利诺伊大学芝加哥分校);University of Texas at Arlington(德克萨斯大学阿灵顿分校)

Suraj Shetiya, Shohedul Hasan, Abolfazl Asudeh, Gautam Das

AI总结:

针对分位数回归(QR)现有结果多为弱多项式、缺乏高效强多项式算法的问题,本文结合几何概念与随机分治策略,提出了适用于二维及高维场景的多种高效强多项式QR算法,给出了各算法的时间复杂度。

摘要:

线性回归是统计学与机器学习领域的一项基础性技术,其目标是在响应(即因)变量与一个或多个预测(即自)变量之间构建线性预测模型。本文重新探讨了经典的分位数回归(Quantile Regression,QR)技术,从统计学角度看,它是另一经典技术普通最小二乘回归(Ordinary Least Square Regression,OLS)的更稳健替代方案。然而,尽管OLS已有高效算法,但目前几乎所有已知的QR相关结果都仅是弱多项式的。为填补这一空白,本文针对多种场景提出了若干用于QR的高效强多项式算法。针对二维QR问题,我们通过建立其与k-集这一几何概念的联系,提出了一种确定性最坏情况时间复杂度为$\boldsymbol{\n^{4/3} polylog(n)}$的算法,其随机化版本的期望时间复杂度为$\boldsymbol{\n^{4/3}}$。我们还提出了一种随机分治算法——RandomizedQR,用于求解二维QR问题的期望时间复杂度为$\boldsymbol{\n\n^2{(n)}}$。对于维度大于2的一般情况,我们的RandomizedQR算法的期望时间复杂度为$\boldsymbol{\n^{d-1}\n^2{(n)}}$。

英文摘要:

Linear Regression is a seminal technique in statistics and machine learning, where the objective is to build linear predictive models between a response (i.e., dependent) variable and one or more predictor (i.e., independent) variables. In this paper, we revisit the classical technique of Quantile Regression (QR), which is statistically a more robust alternative to the other classical technique of Ordinary Least Square Regression (OLS). However, while there exist efficient algorithms for OLS, almost all of the known results for QR are only weakly polynomial. Towards filling this gap, this paper proposes several efficient strongly polynomial algorithms for QR for various settings. For two dimensional QR, making a connection to the geometric concept of $k$-set, we propose an algorithm with a deterministic worst-case time complexity of $\mathcal{O}(n^{4/3} polylog(n))$ and an expected time complexity of $\mathcal{O}(n^{4/3})$ for the randomized version. We also propose a randomized divide-and-conquer algorithm -- RandomizedQR with an expected time complexity of $\mathcal{O}(n\log^2{(n)})$ for two dimensional QR problem. For the general case with more than two dimensions, our RandomizedQR algorithm has an expected time complexity of $\mathcal{O}(n^{d-1}\log^2{(n)})$.

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。

↑