Joint Laplacian feature weights learning

作者：

Highlights：

• Our proposed framework automatically determines the optimal size of the feature subset.

• Our proposed framework imposes nonnegative and l22-norm constraints on feature weights.

• Our proposed framework iteratively learns feature weights jointly and simultaneously.

• Our proposed JLFWL selects the best features corresponding to a given Laplacian graph.

摘要

Highlights•Our proposed framework automatically determines the optimal size of the feature subset.•Our proposed framework imposes nonnegative and l22-norm constraints on feature weights.•Our proposed framework iteratively learns feature weights jointly and simultaneously.•Our proposed JLFWL selects the best features corresponding to a given Laplacian graph.

论文关键词：Feature selection,Joint feature weights learning,Nonnegative,l22-norm

论文评审过程：Received 17 April 2013, Revised 16 August 2013, Accepted 30 September 2013, Available online 14 October 2013.

论文官网地址：https://doi.org/10.1016/j.patcog.2013.09.038