Bayesian Imaging with Data-Driven Priors Encoded by Neural Networks

DOI10.1137/21M1406313MaRDI QIDQ5094622

No author found.

Publication date: 4 August 2022

Published in: SIAM Journal on Imaging Sciences (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2103.10182

zbMATH Keywords

inverse problemsBayesian inferenceMarkov chain Monte Carlo methodsmachine learningmathematical imaging

Mathematics Subject Classification ID

Related Items (3)

Bayesian Inverse Problems Are Usually Well-Posed ⋮ Efficient Bayesian Computation for Low-Photon Imaging Problems ⋮ Enhancing electrical impedance tomography reconstruction using learned half-quadratic splitting networks with Anderson acceleration

Uses Software

• MNIST
• Adam
• RRR
• DnCNN
• BigGAN
• StyleGAN
• PULSE
• DeepAdverserialRegulariser
• NETT

Cites Work

• Unnamed Item
• The Little Engine that Could: Regularization by Denoising (RED)
• A Fast Iterative Shrinkage-Thresholding Algorithm for Linear Inverse Problems
• Spectral gaps for a Metropolis-Hastings algorithm in infinite dimensions
• Towards optimal scaling of Metropolis-coupled Markov chain Monte Carlo
• BigGAN
• Weak convergence and optimal scaling of random walk Metropolis algorithms
• Perturbation theory for Markov chains via Wasserstein distance
• Statistical and computational inverse problems.
• Model distortions in Bayesian MAP reconstruction
• Posterior Expectation of the Total Variation Model: Properties and Experiments
• From Denoising to Compressed Sensing
• Accelerating Proximal Markov Chain Monte Carlo by Using an Explicit Stabilized Method
• Marginal Likelihood Estimation via Power Posteriors
• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• Efficient Bayesian Computation by Proximal Markov Chain Monte Carlo: When Langevin Meets Moreau
• On the Well-posedness of Bayesian Inverse Problems
• NETT: solving inverse problems with deep neural networks
• Vector Approximate Message Passing
• Enhancing Compressed Sensing 4D Photoacoustic Tomography by Simultaneous Motion Estimation
• Benchmark Problems for Phase Retrieval
• Solving inverse problems using data-driven models
• High-Dimensional Mixture Models for Unsupervised Image Denoising (HDMI)
• The Bayesian Choice
• Fast Image Recovery Using Variable Splitting and Constrained Optimization
• An introduction to continuous optimization for imaging
• Testing the manifold hypothesis
• A Stochastic Approximation Method
• Plug in estimation in high dimensional linear inverse problems a rigorous analysis
• Regularization by Denoising via Fixed-Point Projection (RED-PRO)
• MCMC methods for functions: modifying old algorithms to make them faster
• Invariant HPD credible sets and MAP estimators

This page was built for publication: Bayesian Imaging with Data-Driven Priors Encoded by Neural Networks