Feature weight estimation based on dynamic representation and neighbor sparse reconstruction

作者：

Highlights：

• We propose a new dynamic representation framework for feature weight estimation, which redefines the optimization problem.

• Using gradient ascent method, we provide an effective method to solve the optimization problem of DRNSR-Relief and can guarantee its convergence.

• A novel neighbor sparse reconstruction method is proposed for represent neighbors of the given samples.

摘要

•We propose a new dynamic representation framework for feature weight estimation, which redefines the optimization problem.•Using gradient ascent method, we provide an effective method to solve the optimization problem of DRNSR-Relief and can guarantee its convergence.•A novel neighbor sparse reconstruction method is proposed for represent neighbors of the given samples.

论文关键词：Feature weighting,Feature selection,Relief,Sparse learning,Local hyperplane,l1 regularization,Classification

论文评审过程：Received 29 June 2017, Revised 8 January 2018, Accepted 20 March 2018, Available online 13 April 2018, Version of Record 21 April 2018.

论文官网地址：https://doi.org/10.1016/j.patcog.2018.03.014