1: \begin{abstract}

2: 	

3: We propose a novel parametric dictionary learning algorithm for line spectral estimation, applicable in both single measurement vector (SMV) and multiple measurement vectors (MMV) scenarios. 

4: This algorithm, termed {cubic Newtonized K-SVD (NK-SVD)}, extends the traditional K-SVD method by incorporating cubic regularization into Newton refinements.

5: The proposed Gauss-Seidel scheme not only enhances the accuracy of frequency estimation over the continuum but also achieves better convergence by incorporating higher-order derivative information. 

6: We rigorously establish the convergence of the proposed algorithm within the Block Coordinate Descent (BCD) framework. To the best of our knowledge, this is the first convergence analysis of BCD with a higher-order regularization scheme. 

7: Furthermore, this convergence framework can be applied to designing general alternating minimization algorithms with higher-order regularization.

8: Extensive simulations demonstrate that cubic NK-SVD outperforms existing state-of-the-art methods in both SMV and MMV settings.

9: 

10: \end{abstract}

11: