Self-concordant optimization Files

*
* LMAOPT-0.1 is a collection of Matlab code implementing two algorithms proposed in: 
*   [1]. Q. Tran Dinh and Z. Zhang, Extended Gauss-Newton and Gauss-Newton ADMM 
*   algorithms for low-rank matrix optimization, 2016 (Available at: ).
*
* It aims at solving the following low-rank optimization problem:
*                      
*                        min_{U, V} phi(A(UV’) - B),
*
* where phi is a general loss convex function, A is a linear operator, and B is the 
* vector of observations.
*
* In order to install this collection. You should download it from our website and 
* extract it into a given folder and leaves the name of the subfolders unchanged.
*
* If you want to use the mex files, you can compile these three mex files by
*    mex -v GradFxEval.c
*    mex -v ProjUVt.c "-lmwblas"
*
* We have two main solvers: LRMA_LsGnSolver.m and  LRMA_GnAdmmSolver implement the 
* Line-search Gauss-Newton, and Gauss-Newton Alternating direction method of 
* multipliers algorithms, respectively, but specialized to the square-loss function.
* The LRMA_NsGnAdmmSolver implements the Gauss-Newton ADMM algorithm for nonsmooth 
* objective function.
*
* Several examples are provided in Examples subfolder.
* Any feedback or question should be sent to: quoctd@email.unc.edu