1: \begin{abstract}

2:     In this paper, we focus on distributed estimation and support recovery for high-dimensional linear quantile regression. 

3:     Quantile regression is a popular alternative tool to the least squares regression for robustness against outliers and data heterogeneity. 

4:     However, the non-smoothness of the check loss function poses big challenges to both computation and theory in the distributed setting.

5:     To tackle these problems, we transform the original quantile regression into the least-squares optimization. 

6:     By applying a double-smoothing approach, we extend a previous Newton-type distributed approach without the restrictive 

7:     independent assumption between the error term and covariates. 

8:     An efficient algorithm is developed, which enjoys high computation and 

9:     communication efficiency. Theoretically, the proposed distributed estimator achieves a near-oracle convergence rate and high support 

10:     recovery accuracy after a constant number of iterations. 

11:     Extensive experiments on synthetic examples and a real data application further demonstrate the effectiveness of the proposed method.

12: \end{abstract}

13: