1: \begin{abstract}

2: Sparse level-set formulations allow practitioners to find the minimum 1-norm solution subject to likelihood  constraints. Prior art requires this constraint to be convex. In this letter, we develop an efficient approach for nonconvex likelihoods, using Regula Falsi root-finding techniques to solve the level-set formulation.

3: Regula Falsi methods are simple,  derivative-free, 

4: and efficient, and the approach provably extends level-set methods to the broader class of nonconvex inverse problems. 

5: Practical performance is illustrated using $\ell_1$-regularized Student's t inversion, which is a nonconvex approach used to develop outlier-robust formulations. 

6: %In this letter, we propose a new optimality tracing based  noise-aware $\ell_{1}$-norm minimization approach. Different regularization problem formulations can provide equivalent solutions. Using this fact, it is possible to solve one computationally efficient problem to have the solution of another one. 

7: %We solved constrained $\ell_{1}$-norm regularization, and the optimality trade-off between a loss function that measures the data misfit and the $\ell_{1}$-norm regularizer is traced over a pareto frontier with the aim of having the solution of noise-aware $\ell_{1}$-norm regularization. This tracing can be depicted as a nonlinear equation root finding problem. 

8: %Regula Falsi type methods are employed for the root finding steps. 

9: %RegThese methods are simple,  derivative-free, costly efficient, ensure convergence and applicable for the optimization problems with both convex and nonconvex losses.

10: \end{abstract}

11: