Minimax Rates of <i>ℓ</i><sub><i>p</i></sub>-Losses for High-Dimensional Linear Errors-in-Variables Models over <i>ℓ</i><sub><i>q</i></sub>-Balls
In this paper, the high-dimensional linear regression model is considered, where the covariates are measured with additive noise. Different from most of the other methods, which are based on the assumption that the true covariates are fully obtained, results in this paper only require that the corru...
Main Authors: | Xin Li, Dongya Wu |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-06-01
|
Series: | Entropy |
Subjects: | |
Online Access: | https://www.mdpi.com/1099-4300/23/6/722 |
Similar Items
-
Canonical Divergence for Flat <i>α</i>-Connections: Classical and Quantum
by: Domenico Felice, et al.
Published: (2019-08-01) -
Regularization Methods Based on the <i>L<sub>q</sub></i>-Likelihood for Linear Models with Heavy-Tailed Errors
by: Yoshihiro Hirose
Published: (2020-09-01) -
Where Does Minimum Error Entropy Outperform Minimum Mean Square Error? A New and Closer Look
by: Ahmad Reza Heravi, et al.
Published: (2018-01-01) -
Asymptotically Constant-Risk Predictive Densities When the Distributions of Data and Target Variables Are Different
by: Keisuke Yano, et al.
Published: (2014-05-01) -
Theoretical Description of R–X⋯NH<sub>3</sub> Halogen Bond Complexes: Effect of the R Group on the Complex Stability and Sigma-Hole Electron Depletion
by: Juan Zurita, et al.
Published: (2020-01-01)