1: \begin{abstract}

2: %

3:   We study  distributed multi-agent large-scale optimization problems, wherein the cost function is composed of

4: a smooth possibly nonconvex sum-utility  plus  a DC (Difference-of-Convex) regularizer. 

5: We consider the scenario where the dimension of the optimization variables is so large that optimizing and/or transmitting the entire set of variables could cause unaffordable computation and communication overhead. 

6: To address this issue, we propose   the first  distributed algorithm  whereby  agents optimize and communicate   only a portion of their local variables.

7: The scheme hinges on  successive convex approximation (SCA) to handle the nonconvexity of the objective function, coupled with  a novel block-signal tracking scheme, aiming at 

8: locally estimating the average of the agents' gradients.

9:  Asymptotic convergence to

10: stationary solutions of the nonconvex problem is established. Numerical results on  a sparse regression problem show   

11:   the effectiveness of the proposed algorithm and the impact of the block size on its practical convergence speed and communication cost.

12: %

13: \end{abstract}

14: