Unbiased Bregman-Risk Estimators: Application to Regularization Parameter Selection in Tomographic Image Reconstruction
Numerical Analysis
2021-11-22 v2 Numerical Analysis
Optimization and Control
Abstract
Unbiased estimators are introduced for averaged Bregman divergences which generalize Stein's Unbiased (Predictive) Risk Estimator, and the minimization of these estimators is proposed as a regularization parameter selection method for regularization of inverse problems. Numerical experiments are presented in order to show the performance of the proposed technique. Experimental results indicate a useful occurence of a concentration of measure phenomena and some implications of this hypothesis are analyzed.
Cite
@article{arxiv.2109.08784,
title = {Unbiased Bregman-Risk Estimators: Application to Regularization Parameter Selection in Tomographic Image Reconstruction},
author = {Elias S. Helou and Sandra A. Santos and Lucas E. A. Simões},
journal= {arXiv preprint arXiv:2109.08784},
year = {2021}
}