A novel dimension reduction and dictionary learning framework for high-dimensional data classification
作者：
Highlights：
• A novel dimension reduction and dictionary learning framework is proposed.
• At dimension reduction stage, it learns a nonlinear mapping via an autoencoder.
• At dictionary learning stage, it preserves local structure and enhances class discrimination.
• The nonlinear mapping and dictionary are optimized jointly.
• It preserves nonlinear structure within data and results in enhanced classification performance.
摘要
•A novel dimension reduction and dictionary learning framework is proposed.•At dimension reduction stage, it learns a nonlinear mapping via an autoencoder.•At dictionary learning stage, it preserves local structure and enhances class discrimination.•The nonlinear mapping and dictionary are optimized jointly.•It preserves nonlinear structure within data and results in enhanced classification performance.
论文关键词：High-dimensional data classification,Dimension reduction,Dictionary learning,Autoencoder
论文评审过程：Received 16 January 2020, Revised 22 October 2020, Accepted 13 December 2020, Available online 18 December 2020, Version of Record 27 December 2020.
论文官网地址：https://doi.org/10.1016/j.patcog.2020.107793
