1: \begin{abstract}

2: {This paper is concerned with the theoretical understanding of $\alpha$-stable sheets $U$ on $\mathbb{R}^d$ within the framework of Bayesian inference. Our motivation is in the context of Bayesian inverse problems, where we consider the distributions of  random fields $U$  as qualitatively informative prior distributions.  We derive posterior convergence results referring to finite-dimensional approximations of  infinite-dimensional random variables $U$. In doing so we use a number of variants, which these sheets $U$  can take, such as a stochastic integral representation, but also a random series expansion, namely,  the L\'evy-LePage series. Especially,  we study natural discretizations of  $U$, which arise from independent stable  increments. Our proofs rely on the fact  that the random sheets $U$  omit $L^p$-sample paths.  Aside from convergence of approximations of stable sheets, we address whether both well-definedness and well-posedness of the inverse problem can be attained. }

3: 

4: %This paper is concerned with the theoretical understanding of $\alpha$-stable sheets $U$ on $\mathbb{R}^d$ within the framework of Bayesian inference. Our motivation is in the context of Bayesian inverse problems, where we consider the distributions of  random fields $U$  as qualitatively informative prior distributions.  We derive posterior convergence results referring to finite-dimensional approximations of  infinite-dimensional random variables $U$. In doing so we use a number of variants, which these sheets $U$  represent. Our proofs rely on the fact that the random sheets $U$  omit $L^p$-sample paths. We also address whether well-definedness and well-posedness of the inverse problem can be attained.

5: \end{abstract}

6: