1: \begin{abstract}

2: The Ensemble Kalman methodology in an inverse problems setting can be viewed \revcls{\sout{-- when constructed in a sequential Monte-Carlo-like manner --}} {as an iterative scheme, }

3: which  is a weakly tamed discretization scheme 

4: for a certain stochastic differential equation (SDE). \rev{\sout{ The continuous time limit of the method,

5: formally derived by Schillings and Stuart, allows to establish a well-posedness and convergence analysis in the inverse setting.}} 

6: Assuming a suitable approximation result,  

7: dynamical properties of the SDE can be rigorously pulled back via the discrete scheme to the original \revcls{Ensemble Kalman inversion}. 

8: 

9: \rev{The results of this} paper make a step towards closing the gap of the missing approximation result by proving a strong convergence result \rev{in a simplified model of a scalar stochastic differential equation}. 

10: We focus here on a \rev{toy model} with similar properties than the one arising in the \rev{context of} Ensemble Kalman filter. 

11: The proposed model can be interpreted as a single particle filter for a linear map and thus forms the basis for further analysis.

12: The difficulty in the analysis arises from the formally derived limiting SDE with non-globally Lipschitz continuous nonlinearities\rev{ both in the drift and in the diffusion.} 

13: \rev{Here the standard Euler-Maruyama scheme might fail to provi\revcls{d}e a strongly convergent numerical scheme and \revcls{taming} \revcls{is} necessary. 

14: In contrast to the strong \revcls{taming} usually used, the method presented here provides a weaker form of \revcls{taming}.} 

15: 

16: We present a strong convergence analysis by first proving convergence on a domain of high probability \rev{by using a cut-off or localisation}, which then leads, 

17: combined with bounds on moments for both the SDE and the numerical scheme, by a bootstrapping argument to \rev{strong convergence}.

18: \end{abstract}

19: