Sparse exponential family Principal Component Analysis
作者：
Highlights：
• A Sparse exponential family Principal Component Analysis is proposed.
• It has wide applications to any data following exponential family distributions.
• Efficient solutions can be achieved with closed-form update rules.
• The performance is enhanced with appropriate assumption of the data distribution.
• This model is flexible and highly extensible.
摘要
Highlights•A Sparse exponential family Principal Component Analysis is proposed.•It has wide applications to any data following exponential family distributions.•Efficient solutions can be achieved with closed-form update rules.•The performance is enhanced with appropriate assumption of the data distribution.•This model is flexible and highly extensible.
论文关键词：Dimension reduction,Sparsity,Exponential family principal component analysis
论文评审过程：Received 15 January 2016, Revised 9 May 2016, Accepted 19 May 2016, Available online 21 May 2016, Version of Record 13 July 2016.
论文官网地址：https://doi.org/10.1016/j.patcog.2016.05.024
