1: \begin{abstract}

2: This paper proposes two proximal Newton-CG methods for convex nonsmooth optimization problems in composite form.

3: The algorithms are based on a a reformulation of the original nonsmooth problem as

4: the unconstrained minimization of a continuously differentiable function, namely the \emph{forward-backward envelope (FBE)}.

5: The first algorithm is based on a standard line search strategy, whereas the second one 

6: combines the global efficiency estimates of the corresponding first-order methods, while achieving fast asymptotic convergence rates. Furthermore, they are computationally attractive since each Newton iteration requires the approximate solution of a linear system of usually small dimension. 

7: \end{abstract}

8: