学术报告
丁宏强教授:Recent Advances in L1, L21, L12 Research

 

Academy of Mathematics and Systems Science, CAS
Colloquia & Seminars

Speaker:

丁宏强教授,香港中文大学(深圳)

Inviter: 张世华 研究员
Title:
Recent Advances in L1, L21, L12 Research
Language: Chinese
Time & Venue:
2023.03.20 10:30-11:30 南楼226
Abstract:

We report 3 recent works along this direction: (A) The nature of L1, L21 error function robustness, using an outlier regularization framework.(B)Labeled robust principal component analysis (RPCA), by adding class label information into the popular RPCA, can significantly improve the robust data recovery. (C) Using L12 norm for a new different type of feature selection.