1: \begin{abstract}

2: 		In this paper we study nonconvex and nonsmooth optimization problems with semi-algebraic data, 

3: 		where the variables vector is split into several blocks of variables. The problem consists of 

4: 		one smooth function of the entire variables vector and the sum of nonsmooth functions for each 

5: 		block separately. We analyze an inertial version of the Proximal Alternating Linearized 

6: 		Minimization (PALM) algorithm and prove its global convergence to a critical point of the 

7: 		objective function at hand. We illustrate our theoretical findings by presenting numerical 

8: 		experiments on blind image deconvolution, on sparse non-negative matrix factorization and on 

9: 		dictionary learning, which demonstrate the viability and effectiveness of the proposed method.

10:     \end{abstract}

11: