1: \begin{abstract}

2: 

3: Non-convex sparse regularizers are common tools for learning with high-dimensional

4: data. For accelerating convergence of a Lasso problem using those

5: regularizers, a working set strategy addresses the optimization problem through an iterative

6: algorithm by gradually incrementing the number of variables to optimize until the

7: identification of the solution support. We propose in this paper the first Lasso working set algorithm 

8: for non-convex sparse regularizers with convergence guarantees.

9: The algorithm, named \emph{FireWorks}, is based on a non-convex reformulation of

10: a recent  duality-based approach  and leverages on the geometry of the

11: residuals. We provide theoretical guarantees 

12: showing that convergence is preserved even when the inner solver is inexact,

13: under sufficient decay of the error across iterations.

14: Experimental results demonstrate strong computational gain when using our working set strategy compared to full problem solvers for both block-coordinate descent or a proximal gradient solver.

15: \end{abstract}

16: