Regularization independent of the noise level: an analysis of quasi-optimality

Abstract

The quasi-optimality criterion chooses the regularization parameter in inverse problems without taking into account the noise level. This rule works remarkably well in practice, although Bakushinskii has shown that there are always counterexamples with very poor performance. We propose an average case analysis of quasi-optimality for spectral cut-off estimators and we prove that the quasi-optimality criterion determines estimators which are rate-optimal {\em on average}. Its practical performance is illustrated with a calibration problem from mathematical finance.

18 pages, 3 figures

Cited by in corpus (10)

• Statistical inference for time-changed Lévy processes via composite characteristic function estimation
• Beyond the Bakushinskii veto: Regularising linear inverse problems without knowing the noise distribution
• Supplementary Appendix for: Constrained Perturbation Regularization Approach for Signal Estimation Using Random Matrix Theory
• Estimation of the Regularization Parameter in Linear Discrete Ill-Posed Problems Using the Picard parameter
• Optimal rates for the regularized learning algorithms under general source condition
• A modified discrepancy principle to attain optimal convergence rates under unknown noise
• Regularising linear inverse problems under unknown non-Gaussian white noise allowing repeated measurements
• On Learning the Optimal Regularization Parameter in Inverse Problems
• Concentration Inequalities for Cross-validation in Scattered Data Approximation
• Dual-grid parameter choice method with application to image deblurring