Efficient Bayesian Computation Using Plug-and-Play Priors for Poisson Inverse Problems
Not on the map. Why?
Abstract (arXiv)
This paper studies plug-and-play (PnP) Langevin sampling strategies for Bayesian inference in low-photon Poisson imaging problems, a challenging class of problems with significant applications in astronomy, medicine, and biology. PnP Langevin sampling offers a powerful framework for Bayesian image restoration, enabling accurate point estimation as well as advanced inference tasks, including uncertainty quantification and visualization analyses, and empirical Bayesian inference for automatic model parameter tuning. Herein, we leverage and adapt recent developments in this framework to tackle challenging imaging problems involving weakly informative Poisson data. Existing PnP Langevin algorithms are not well-suited for low-photon Poisson imaging due to high solution uncertainty and poor regularity properties, such as exploding gradients and non-negativity constraints. To address these challenges, we explore two strategies for extending Langevin PnP sampling to Poisson imaging models: (i) an accelerated PnP Langevin method that incorporates boundary reflections and a Poisson likelihood approximation and (ii) a mirror sampling algorithm that leverages a Riemannian geometry to handle the constraints and the poor regularity of the likelihood without approximations. The effectiveness of these approaches is evaluated and contrasted through extensive numerical experiments and comparisons with state-of-the-art methods. The source code accompanying this paper is available at https://github.com/freyyia/pnp-langevin-poisson.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 43 of its 83 references that are in the map. First posted on arXiv 20 Mar 2025.
Builds on
43 of 83 references.
References · 43
- 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
- 2016Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoisingcited by 1,358
- 2020Plug-and-Play Image Restoration with Deep Denoiser Priorcited by 470
- 2016The Little Engine that Could: Regularization by Denoising (RED)cited by 307
- 2014Theoretical guarantees for approximate sampling from smooth and log-concave densitiescited by 208
- 2019Plug-and-Play Methods Provably Converge with Properly Trained Denoiserscited by 198
- 2015Non-asymptotic convergence analysis for the Unadjusted Langevin Algorithmcited by 197
- 2021Gradient Step Denoiser for convergent Plug-and-Playcited by 108
- 2021Bayesian imaging using Plug & Play priors: when Langevin meets Tweediecited by 97
- 2016Efficient Bayesian computation by proximal Markov chain Monte Carlo: when Langevin meets Moreaucited by 84
- 2022Proximal Denoiser for Convergent Plug-and-Play Optimization with Nonconvex Regularizationcited by 80
- 2018Sharp convergence rates for Langevin dynamics in the nonconvex settingcited by 68
31 more
- 2020Learning Maximally Monotone Operators for Image Recoverycited by 62
- 2018Mirrored Langevin Dynamicscited by 61
- 2021Robust Equivariant Imaging: a fully unsupervised framework for learning to image from noisy and partial measurementscited by 56
- 2020Wasserstein Control of Mirror Langevin Monte Carlocited by 55
- 2020On the Convergence of Langevin Monte Carlo: The Interplay between Tail Growth and Smoothnesscited by 55
- 2020Efficient constrained sampling via the mirror-Langevin algorithmcited by 55
- 2023Provable Probabilistic Imaging using Score-Based Generative Priorscited by 51
- 2022Learned reconstruction methods with convergence guaranteescited by 46
- 2023Plug-and-Play split Gibbs sampler: embedding deep generative priors in Bayesian inferencecited by 42
- 2022A Neural-Network-Based Convex Regularizer for Inverse Problemscited by 41
- 2021The Mirror Langevin Algorithm Converges with Vanishing Biascited by 37
- 2023Learning Weakly Convex Regularizers for Convergent Image-Reconstruction Algorithmscited by 31
- 2019Langevin Monte Carlo without smoothnesscited by 28
- 2021The Langevin Monte Carlo algorithm in the non-smooth log-concave casecited by 26
- 2023Equivariant plug-and-play image reconstructioncited by 26
- 2019A Continuous-time Perspective for Modeling Acceleration in Riemannian Optimizationcited by 25
- 2023Score-Based Generative Models for PET Image Reconstructioncited by 22
- 2015Poisson Inverse Problems by the Plug-and-Play schemecited by 17
- 2024UNSURE: self-supervised learning with Unknown Noise level and Stein's Unbiased Risk Estimatecited by 16
- 2023Provably Convergent Plug-and-Play Quasi-Newton Methodscited by 15
- 2023Convergent Bregman Plug-and-Play Image Restoration for Poisson Inverse Problemscited by 14
- 2023Convergent regularization in inverse problems and linear plug-and-play denoiserscited by 13
- 2021Photon Limited Non-Blind Deblurring Using Algorithm Unrollingcited by 10
- 2023Plug-and-Play Posterior Sampling under Mismatched Measurement and Prior Modelscited by 8
- 2022Bregman Proximal Langevin Monte Carlo via Bregman--Moreau Envelopescited by 7
- 2022SwISS: A Scalable Markov chain Monte Carlo Divide-and-Conquer Strategycited by 6
- 2023Scalable Bayesian uncertainty quantification with data-driven priors for radio interferometric imagingcited by 6
- 2024Empirical Bayesian image restoration by Langevin sampling with a denoising diffusion implicit priorcited by 6
- 2023NF-ULA: Langevin Monte Carlo with Normalizing Flow Prior for Imaging Inverse Problemscited by 4
- 2024Unsupervised Training of Convex Regularizers using Maximum Likelihood Estimationcited by 3
- 2024Accelerated optimization algorithms and ordinary differential equations: the convex non Euclidean casecited by 1
Built on by
Citing papers per year: 2025: 2.
2 papers.