Applies Alternating Direction Method of Multipliers to the l1-regularized quadratic program $$f(\mathbf{x}) + g(\mathbf{x}) = \frac{1}{2}\mathbf{x}^TA\mathbf{x} - d^T\mathbf{x} + \lambda |\mathbf{x}|_1$$
ADMM_EN2(R, d, x0, lam, mu, maxits, tol, quiet)Upper triangular matrix in Chol decomp \(\mu I + A = R^T R\).
nx1 dimensional column vector.
Regularization parameter for l1 penalty, must be greater than zero.
Augmented Lagrangian penalty parameter, must be greater than zero.
Number of iterations to run
Vector of stopping tolerances, first value is absolute, second is relative tolerance.
Logical controlling display of intermediate statistics.
Step length.
ADMM_EN2 returns an object of class "ADMM_EN2" including a list
with the following named components
callThe matched call.
xFound solution.
yDual solution.
zSlack variables.
kNumber of iterations used.
This function is used by other functions and should only be called explicitly for debugging purposes.
Used by: SDAD and the SDADcv cross-validation version.