Joint feature weighting and adaptive graph-based matrix regression for image supervised feature Selection
作者：
Highlights：
• Take image data as input in regression model to keep the spatial relations of elements in data.
• Use the learned feature weight matrix to select important features from image.
• Adaptively learn graph matrix to reduce the influence of noises and preserve the local structure of samples.
• Design an iterative optimization algorithm and analyze its complexity.
• Verify the superiority of the proposed method on several datasets.
摘要
•Take image data as input in regression model to keep the spatial relations of elements in data.•Use the learned feature weight matrix to select important features from image.•Adaptively learn graph matrix to reduce the influence of noises and preserve the local structure of samples.•Design an iterative optimization algorithm and analyze its complexity.•Verify the superiority of the proposed method on several datasets.
论文关键词：Matrix regression,Feature selection,Feature weight matrix,Graph matrix,Classification
论文评审过程：Received 7 March 2020, Revised 18 October 2020, Accepted 25 October 2020, Available online 4 November 2020, Version of Record 10 November 2020.
论文官网地址：https://doi.org/10.1016/j.image.2020.116044
