1: \begin{abstract}

2: Reference \cite{Siop15} investigated the almost sure weak

3: convergence of block-coordinate fixed point algorithms and

4: discussed their applications to nonlinear analysis and optimization.

5: This algorithmic framework features random sweeping rules to select

6: arbitrarily

7: the blocks of variables that are activated over the course of the

8: iterations and it allows for stochastic errors in the evaluation of

9: the operators. The present paper establishes results on the

10: mean-square and linear convergence of the iterates. Applications to

11: monotone operator splitting and proximal optimization algorithms

12: are presented.  

13: \end{abstract}