L1/2 regularized logistic regression

[Mark_L_Stone]

Your output does not show a claim that the (failed) problem is infeasible. The output shoiws a failure of CVX’s successive approximation method, which can fail even for problems which are well formulated and have (finite) optimal solutions.

Use CVXQUAD

CVXQUAD: How to use CVXQUAD's Pade Approximant instead of CVX's unreliable Successive Approximation for GP mode, log, exp, entr, rel_entr, kl_div, log_det, det_rootn, exponential cone. CVXQUAD's Quantum (Matrix) Entropy & Matrix Log related functions

How to use CVXQUAD’s better performing and more reliable Pade Approximant instead of CVX’s unreliable Successive Approximation method: Update as of the release of CVX 2.2 in January 2020: Except for use of Quantum (Matrix) Entropy and Matrix Logarithm related functions addressed later in this post, none of the following, to include reformulation, is necessary if CVX 2.2 is used with Mosek 9.x or higher as solver, in which case CVX will utilize Mosek’s native exponential cone capability, thereby…

and show us the results. I do not guarantee the problem will be solved using CVXQUAD, but it might. In order to maximize your chances of success, if you have CVX professkional or academic versions, and have Gurobi or Mosek solvers available, specify one of them as cvx_solver, because they are better solvers than SeDuMi and SDPT3.