Reconciling ''priors'' and ''priors'' without prejudice ?
Abstract
We discuss a long-lasting {\em qui pro quo} between regularization-based and Bayesian-based approaches to inverse problems, and review some recent results that try to reconcile both viewpoints. This sheds light on some tradeoff between computational efficiency and estimation accuracy in sparse regularization.
Origin | Files produced by the author(s) |
---|
Loading...