1: \begin{abstract}

2: We derive and analyse a new variant of the iteratively regularized Landweber iteration, for solving linear and nonlinear 

3: ill-posed inverse problems. The method takes into account training 

4: data, which are used to estimate the interior of a black box, which is used to define the iteration process.  

5: We prove convergence and stability for the scheme in infinite dimensional Hilbert spaces. 

6: These theoretical results are complemented by several numerical experiments for solving linear inverse problems for the Radon 

7: transform and a nonlinear inverse problem for Schlieren tomography.

8: %, taking as damping term a linear map between finite dimensional spaces. 

9: \end{abstract}

10: