1: \begin{abstract}
2: Backward stochastic differential equations (BSDEs)
3: appear in numeruous applications.
4: Classical approximation methods suffer from the curse of dimensionality
5: and deep learning-based approximation methods are not known to converge to the BSDE solution.
6: Recently, Hutzenthaler et al.\ \cite{hutzenthaler2021overcoming} introduced a new
7: approximation method for BSDEs whose forward diffusion is Brownian motion
8: and proved that this method converges with essentially optimal rate without suffering from
9: the curse of dimensionality.
10: The central object of this article is to extend this result to general forward diffusions.
11: The main challenge is that we need to establish convergence in temporal-spatial H\"older norms
12: since the forward diffusion cannot be sampled exactly in general.
13: \end{abstract}
14: