1: \begin{abstract}

2: This paper describes a new MATLAB software package of iterative regularization

3: methods and test problems for large-scale linear inverse problems.

4: The software package, called \IRT, serves two related purposes:\

5: we provide implementations of a range of iterative solvers,

6: including several recently proposed methods that are not

7: available elsewhere, and we provide a set of large-scale test

8: problems in the form of discretizations of 2D linear inverse problems.

9: The solvers include iterative regularization methods where the

10: regularization is due to the semi-convergence of the iterations,

11: Tikhonov-type formulations where the regularization is explicitly

12: formulated in the form of a regularization term, and methods that can impose

13: bound constraints on the computed solutions.

14: All the iterative methods are implemented in a very flexible fashion

15: that allows the problem's coefficient matrix to be available as

16: a (sparse) matrix, a function handle, or an object.

17: The most basic call to all of the various iterative methods requires

18: only this matrix and the right hand side vector; if the method

19: uses any special stopping criteria, regularization parameters, etc.,

20: then default values are set automatically by the code. Moreover,

21: through the use of an optional input structure,

22: the user can also have full control of any of the algorithm parameters.

23: The test problems represent realistic large-scale problems found

24: in image reconstruction and several other applications.  Numerical

25: examples illustrate the various algorithms and test problems

26: available in this package.

27: \end{abstract}

28: