Wasserstein-based Projections with Applications to Inverse Problems
Preprint
Abstract
Inverse problems consist of recovering a signal from a collection of noisy measurements. These are typically cast as optimization problems, with classic approaches using a data fidelity term and an analytic regularizer that stabilizes recovery. Recent Plug-and-Play (PnP) works propose replacing the operator for analytic regularization in optimization methods by a data-driven denoiser. These schemes obtain state of the art results, but at the cost of limited theoretical guarantees. To bridge this gap, we present a new algorithm that takes samples from the manifold of true data as input and outputs an approximation of the projection operator onto this manifold. Under standard assumptions, we prove this algorithm generates a learned operator, called Wasserstein-based projection (WP), that approximates the true projection with high probability. Thus, WPs can be inserted into optimization methods in the same manner as PnP, but now with theoretical guarantees. Provided numerical examples show WPs obtain state of the art results for unsupervised PnP signal recovery.Keywords
All Related Versions
This publication has 54 references indexed in Scilit:
- A two-stage method for spectral–spatial classification of hyperspectral imagesJournal of Mathematical Imaging and Vision, 2020
- Deep Neural Networks Motivated by Partial Differential EquationsJournal of Mathematical Imaging and Vision, 2019
- Learning a variational network for reconstruction of accelerated MRI dataMagnetic Resonance in Medicine, 2017
- Convex Analysis and Monotone Operator Theory in Hilbert Spaces2017
- Iterative Methods for Fixed Point Problems in Hilbert Spaces2013
- A First-Order Primal-Dual Algorithm for Convex Problems with Applications to ImagingJournal of Mathematical Imaging and Vision, 2010
- Best Choices for Regularization Parameters in Learning Theory: On the Bias—Variance ProblemMathematische Annalen, 2002
- Best Approximation in Inner Product Spaces2001
- Fast Simulation of 3D Electromagnetic Problems Using PotentialsJournal of Computational Physics, 2000
- Ishikawa and Mann Iterative Process with Errors for Nonlinear Strongly Accretive Mappings in Banach SpacesJournal of Mathematical Analysis and Applications, 1995