Spectral Projected Gradient Method for the Positive Semi-definite Procrustes Problem

This file contains a spectral projected gradiente method for PSDP.

---------------------------------------------------------------------------------
Description:
This code implements an algorithm to solve the PSD Procrustes problem:
given rectangular matrices A and B, find the symmetric positive
semidefinite matrix X that minimizes the Frobenius norm of XA-B, i.e.
min 0.5||XA-B||_F^{2} s.t. X\in S_{+}(n),

where S_{+}(n) denote the set form by the all symmetric and
positive semi-definite matrices of size n-by-n with real entries.
---------------------------------------------------------------------------------

Reference:
Harry F. Oviedo Leon
"Un Método de Gradiente Projectado Espectral para el Problema
Procrustes Semidefinido Positivo". (ResearchGate)

In english: "Spectral Projected Gradient Method for the Positive
Semi-definite Procrustes Problem".

Author: Harry F. Oviedo Leon

Date: 01-Oct-2017

See help

>> help demo_PSDP:

% Example 1:
>> demo_PSDP()

% Example 2:
>> demo_PSDP(1000,1000,1,0)