Risk hull method for spectral regularization in linear statistical inverse problems
We consider in this paper the statistical linear inverse problem = + where denotes a compact operator, a noise level and a stochastic noise. The unknown function has to be recovered from the indirect measurement . We are interested in the following approach: given a family of estimators, we want to select the best possible one. In this context, the unbiased risk estimation (URE) method is rather popular. Nevertheless, it is also very unstable. Recently, Cavalier and Golubev...