Multi-label feature selection via robust flexible sparse regularization

作者：

Highlights：

• A regularization norm named robust flexible sparse regularization (RFSR) is designed.

• The proposed norm overcomes the limitations of existing norms.

• RFSR is introduced into the proposed framework with the global solution.

• An efficient optimization method to solve the proposed method is designed.

摘要

•A regularization norm named robust flexible sparse regularization (RFSR) is designed.•The proposed norm overcomes the limitations of existing norms.•RFSR is introduced into the proposed framework with the global solution.•An efficient optimization method to solve the proposed method is designed.

论文关键词：Multi-label learning,Feature selection,Sparse regularization,Classification

论文评审过程：Received 24 November 2021, Revised 6 July 2022, Accepted 25 September 2022, Available online 27 September 2022, Version of Record 30 September 2022.

论文官网地址：https://doi.org/10.1016/j.patcog.2022.109074