1: \begin{abstract}

2:   We study linear rough partial differential equations in the setting of [Friz

3:   and Hairer, Springer, 2014, Chapter 12]. More precisely, we consider a

4:   linear parabolic partial differential equation driven by a deterministic

5:   rough path $\mathbf{W}$ of H\"{o}lder regularity $\alpha$ with $1/3 < \alpha

6:   \le 1/2$.  Based on a stochastic representation of the solution of the rough

7:   partial differential equation, we propose a regression Monte Carlo algorithm

8:   for spatio-temporal approximation of the solution.  We provide a full

9:   convergence analysis of the proposed approximation method which essentially

10:   relies on the new bounds for the higher order derivatives of the solution in

11:   space. Finally, a comprehensive simulation study showing the applicability

12:   of the proposed algorithm is presented.

13: \end{abstract}