April 2018

This subdirectory contains the following Matlab source codes for the simultaneous sparse recovery and outlier detection problem with trimmed least squares and truncated l1 regularizer:

p_DCAe.m                 Applying pDCAe to the DC reformulation 

NPG_DC.m                 Applying NPG with Majorization to the DC reformulation

Run_test_new.m           run codes.

Proj.m, soft_thresh.m    Subroutine for pDCAe and NPG_DC

Implementation and numerical experience with the above codes are described in the paper: 
    Tianxiang Liu, Ting Kei Pong and Akiko Takeda
    "A refined convergence analysis of pDCAe with applications to simultaneous sparse recovery and outlier detection",
    Submitted.
This code was last updated on April 12, 2018.

Questions/comments/suggestions about the codes are welcome.  

Ting Kei Pong tk.pong@polyu.edu.hk