1: \begin{abstract}

2: We study a stochastically perturbed version of the well-known Krasnoselski--Mann iteration for  computing fixed points of nonexpansive 

3: maps in finite dimensional normed spaces.

4: We discuss sufficient conditions on the stochastic noise and stepsizes that guarantee almost sure convergence of the iterates towards 

5: a fixed point, and derive non-asymptotic error bounds and convergence rates for the fixed-point residuals. 

6: Our main results concern the case of a martingale difference noise with variances that can possibly grow unbounded.

7: This supports an application to reinforcement learning for average reward Markov decision processes, for which we establish

8: convergence and asymptotic rates. We also 

9: analyze in depth the case where the noise has uniformly bounded variance, obtaining error bounds 

10: with explicit computable constants. 

11:  \end{abstract}

12: