1: \begin{abstract} 

2: Dictionary Learning (DL) is one of the leading

3: sparsity promoting techniques in the context of image classification, where the ``dictionary'' matrix

4: $D$ of images and the sparse matrix $X$ are determined so as to represent a redundant image dataset. 

5: The resulting constrained optimization problem is nonconvex and non-smooth, providing several

6: computational challenges for its solution.

7: To preserve multidimensional data features, various tensor DL formulations 

8: have been introduced, adding to the problem complexity. 

9: We develop a new alternating algorithm for the solution of

10: the DL problem both in the matrix and tensor frameworks; in the latter case

11: a new formulation based on Tensor-Train decompositions is also proposed.

12: The new method belongs to the 

13: Proximal Alternating Linearized Minimization (PALM) algorithmic family, with the inclusion of

14: second order information to enhance efficiency. We discuss a rigorous convergence analysis, and

15: report on the new method performance on the image classification of several benchmark datasets.

16: 

17: \end{abstract}

18: