PreprintarXiv licencearXiv
Data-driven Smooth-to-hard Spectral Regularization for General Linear Inverse Problems
Abstract
The logistic smooth-to-hard shrinker is the proximal map of an explicit saturating spectral penalty for general linear inverse problems. Its exact weak convexity yields a unique scaled update and proximal-gradient stationarity guarantees, while independent validation provides finite-grid parameter-selection control. We use MRI as an illustrative example.
Related
- Data-driven stochastic asymptotic regularization for ill-posed inverse problems with a posteriori and heuristic stopping
- Data-driven Regularization for Inverse Problems
- Physics-Guided Conditional Flow Matching with Energy Regularization for Robust PDE Inverse Problems
- A Spectral Model-Informed Neural Network for Inverse Source Problems
- Finite inverse nodal problems for singular weighted Sturm--Liouville equations: variational selection and spectral matching
BibTeX
@misc{yang2026datadriven,
title = {{Data-driven Smooth-to-hard Spectral Regularization for General Linear Inverse Problems}},
author = {Guanzhong Yang and Zi Wang and Guang Yang},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.21664},
eprint = {2610.21664},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.21664},
}
RIS
TY - UNPB
TI - Data-driven Smooth-to-hard Spectral Regularization for General Linear Inverse Problems
AU - Guanzhong Yang
AU - Zi Wang
AU - Guang Yang
PY - 2026
DA - 2026/10/15
PB - arXiv
DO - 10.48550/arxiv.2610.21664
UR - https://arxiv.org/abs/2610.21664
AB - The logistic smooth-to-hard shrinker is the proximal map of an explicit saturating spectral penalty for general linear inverse problems. Its exact weak convexity yields a unique scaled update and proximal-gradient stationarity guarantees, while independent validation provides finite-grid parameter-selection control. We use MRI as an illustrative example.
ER -