Wasserstein-Based Projections with Applications to Inverse Problems

DOI10.1137/20M1376790zbMath1492.94019arXiv2008.02200OpenAlexW3156082199MaRDI QIDQ5074785

Wotao Yin, Samy Wu Fung, Alex Tong Lin, Howard Heaton, Stanley J. Osher

Publication date: 10 May 2022

Published in: SIAM Journal on Mathematics of Data Science (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2008.02200

zbMATH Keywords

computed tomographyinverse problemsprojectiondeep neural networkplug-and-playgenerative adversarial networkWassersteingenerative modelinglearning to optimizeHalpern

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Ill-posedness and regularization problems in numerical linear algebra (65F22) Numerical mathematical programming methods (65K05) Convex programming (90C25) Biomedical imaging and signal processing (92C55) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08)

Related Items

Fast sinkhorn. II: Collinear triangular matrix and linear time accurate computation of optimal transport, Deep unfolding as iterative regularization for imaging inverse problems, An Optimal Transport Analogue of the Rudin–Osher–Fatemi Model and Its Corresponding Multiscale Theory

Uses Software

• PyTorch
• Adam
• t-SNE
• ODL
• GitHub
• PhaseLift
• ADMM-Softmax
• U-Net
• PNKH-B
• Wasserstein GAN
• OT-Flow
• DeepAdverserialRegulariser
• Entropy-SGD
• LoDoPaB-CT

Cites Work

• Unnamed Item
• Unnamed Item
• Nonlinear total variation based noise removal algorithms
• A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems
• Iterative methods for fixed point problems in Hilbert spaces
• On the local behavior of spaces of natural images
• The nonlinear statistics of high-contrast patches in natural images
• Tikhonov regularization of large linear problems
• Fast simulation of 3D electromagnetic problems using potentials.
• A multiscale method for model order reduction in PDE parameter estimation
• Best choices for regularization parameters in learning theory: on the bias-variance problem.
• A first-order primal-dual algorithm for convex problems with applications to imaging
• Deep neural networks motivated by partial differential equations
• ADMM-softmax: an ADMM approach for multinomial logistic regression
• Deep neural network structures solving variational inequalities
• A two-stage method for spectral-spatial classification of hyperspectral images
• Quantitative robust uncertainty principles and optimally sparse decompositions
• A General Framework for a Class of First Order Primal-Dual Algorithms for Convex Optimization in Imaging Science
• Image Processing with Nonlocal Spectral Bases
• Large-Scale Machine Learning with Stochastic Gradient Descent
• Deblurring Images
• Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information
• Optical tomography: forward and inverse problems
• Optical tomography in medical imaging
• Solving ill-posed inverse problems using iterative deep neural networks
• Stable architectures for deep neural networks
• Deep Convolutional Neural Network for Inverse Problems in Imaging
• Low Dimensional Manifold Model for Image Processing
• Tikhonov Regularization and Total Least Squares
• Laplacian Eigenmaps for Dimensionality Reduction and Data Representation
• Phase Retrieval with Sparse Phase Constraint
• Lipschitz Certificates for Layered Network Structures Driven by Averaged Activation Operators
• Multigrid Optimization for Large-Scale Ptychographic Phase Retrieval
• Computed tomography reconstruction using deep image prior and learned reconstruction methods
• PNKH-B: A Projected Newton--Krylov Method for Large-Scale Bound-Constrained Optimization
• An Uncertainty-Weighted Asynchronous ADMM Method for Parallel PDE Parameter Estimation
• An Accelerated Linearized Alternating Direction Method of Multipliers
• A Computational Framework for Infinite-Dimensional Bayesian Inverse Problems Part I: The Linearized Case, with Application to Global Seismic Inversion
• Bregman Iterative Algorithms for $\ell_1$-Minimization with Applications to Compressed Sensing
• Proximité et dualité dans un espace hilbertien
• Fixed points of nonexpanding maps
• An Iterative Regularization Method for Total Variation-Based Image Restoration
• A Stochastic Approximation Method
• Entropy-SGD: biasing gradient descent into wide valleys
• Regularization by Denoising via Fixed-Point Projection (RED-PRO)
• Learning Filter Functions in Regularisers by Minimising Quotients
• Phase Retrieval via Matrix Completion
• Optimal Transport
• Compressed sensing