1: \begin{abstract}

2: We consider the Bayesian approach to linear inverse problems when

3: the underlying operator depends on an unknown parameter. Allowing

4: for finite dimensional as well as infinite dimensional parameters,

5: the theory covers several models with different levels of uncertainty

6: in the operator. Using product priors, we prove contraction rates

7: for the posterior distribution which coincide with the optimal convergence

8: rates up to logarithmic factors. In order to adapt to the unknown

9: smoothness, an empirical Bayes procedure is constructed based on Lepski's

10: method. The procedure is illustrated in numerical examples.

11: \end{abstract}

12: