Comment on “Joint sparse principal component analysis” by S. Yi et al. (Pattern Recognition, vol. 61, pp. 524–536, 2017)
作者：
Highlights：
• This paper comments on the published work dealing with “Joint sparse principal component analysis” (Pattern Recognition, vol. 61, pp. 524–536, 2017) proposed by S. Yi et al.
• Joint sparse principal component analysis (JSPCA) was proposed to jointly select useful features and enhance robustness to outliers.
• This approach is based on a mathematical model.
• S. Yi et al. proposed a theorem to show that the approach converges to a local optimal solution.
• In this paper, their proof is rejected.
摘要
•This paper comments on the published work dealing with “Joint sparse principal component analysis” (Pattern Recognition, vol. 61, pp. 524–536, 2017) proposed by S. Yi et al.•Joint sparse principal component analysis (JSPCA) was proposed to jointly select useful features and enhance robustness to outliers.•This approach is based on a mathematical model.•S. Yi et al. proposed a theorem to show that the approach converges to a local optimal solution.•In this paper, their proof is rejected.
论文关键词：Joint sparse principal component analysis (JSPCA),Feature selection,Convergence,Local optimal solution
论文评审过程：Received 25 November 2016, Accepted 8 October 2017, Available online 10 October 2017, Version of Record 6 February 2018.
论文官网地址：https://doi.org/10.1016/j.patcog.2017.10.010
