On Convergence Rates of Convex Regression in Multiple Dimensions

On Convergence Rates of Convex Regression in Multiple Dimensions
复制标题

多维凸回归的收敛率

DOI:
10.1287/ijoc.2013.0587
复制
发表时间:
2014
期刊:
INFORMS J. Comput.
影响因子:
--
通讯作者:
Eunji Lim
Eunji Lim
中科院分区:
--
文献类型:
--
作者:
Eunji Lim

文献摘要

被引文献

相似文献

我们考虑一个最小二乘估计估计凸函数f *: [0,1] d(右)(R-openface)具有有界的次梯度。计算估计量与真函数f *之间的平方差之和收敛于零的速率。这项工作对计算多维凸回归估计量的收敛速度有一定的启发。
We consider a least squares estimator for estimating a convex function f * : [0, 1] d (rightarrow) (R-openface) with bounded subgradients. A rate at which the sum of squared differences between the estimator and the true function f * converges to zero is computed. This work sheds light on computing the convergence rate of the multidimensional convex regression estimator.