Answers are grounded in this paper and cite their sources.
We study two nonlinear methods for statistical linear inverse problems when the operator is not known. The two constructions combine Galerkin regularization and wavelet thresholding. Their performances depend on the underlying structure of the operator, quantified by an index of sparsity. We prove their rate-optimality and adaptivity properties over Besov classes.
No takes yet. Share an insight, caveat, or question.
A 2008 study studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: