Unified distributed robust regression and variable selection framework for massive data

作者：

Highlights：

• Unified distributed robust regression framework for massive data is proposed.

• The new method can be easily implemented on the master machine.

• The communication cost is significantly reduced.

• Theoretical properties are established under mild conditions.

摘要

•Unified distributed robust regression framework for massive data is proposed.•The new method can be easily implemented on the master machine.•The communication cost is significantly reduced.•Theoretical properties are established under mild conditions.

论文关键词：Distributed massive data,Robust regression,Communication efficiency,Variable selection

论文评审过程：Received 19 December 2020, Revised 17 June 2021, Accepted 30 July 2021, Available online 6 August 2021, Version of Record 9 August 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115701