mlsanityData to 7 Oct 2026

Scalable Plug-and-Play ADMM with Convergence Guarantees

arXiv 2006.03224 ↗First posted 5 Jun 2020cs.LG math.OC stat.MLcited by 49

Not on the map. Why?

Abstract (arXiv)

Plug-and-play priors (PnP) is a broadly applicable methodology for solving inverse problems by exploiting statistical priors specified as denoisers. Recent work has reported the state-of-the-art performance of PnP algorithms using pre-trained deep neural nets as denoisers in a number of imaging applications. However, current PnP algorithms are impractical in large-scale settings due to their heavy computational and memory requirements. This work addresses this issue by proposing an incremental variant of the widely used PnP-ADMM algorithm, making it scalable to large-scale datasets. We theoretically analyze the convergence of the algorithm under a set of explicit assumptions, extending recent theoretical results in the area. Additionally, we show the effectiveness of our algorithm with nonsmooth data-fidelity terms and deep neural net priors, its fast convergence compared to existing PnP algorithms, and its scalability in terms of speed and memory.

Key facts. Cited by 49 papers in the map (data through 7 Oct 2026). First posted on arXiv 5 Jun 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 5; 2022: 14; 2023: 11; 2024: 5; 2025: 10; 2026: 4.

49 papers.

Sort:

By area:

  1. 2021Learning to Optimize: A Primer and A Benchmarkcited by 206
  2. 2021Gradient Step Denoiser for convergent Plug-and-Playcited by 107
  3. 2022Proximal Denoiser for Convergent Plug-and-Play Optimization with Nonconvex Regularizationcited by 79
  4. 2023Image Denoising: The Deep Learning Revolution and Beyond -- A Survey Paper --cited by 71
  5. 2023Provable Probabilistic Imaging using Score-Based Generative Priorscited by 51
  6. 2021Recovery of Continuous 3D Refractive Index Maps from Discrete Intensity-Only Measurements using Neural Fieldscited by 48
  7. 2023What's in a Prior? Learned Proximal Networks for Inverse Problemscited by 24
  8. 2022Improving Lipschitz-Constrained Neural Networks by Learning Activation Functionscited by 23
  9. 2021Recovery Analysis for Plug-and-Play Priors using the Restricted Eigenvalue Conditioncited by 22
  10. 2022Online Deep Equilibrium Learning for Regularization by Denoisingcited by 22
  11. 2022Theoretical Perspectives on Deep Learning Methods in Inverse Problemscited by 22
  12. 2024Learning Image Priors through Patch-based Diffusion Models for Solving Inverse Problemscited by 18
37 more
  1. 2023Provably Convergent Plug-and-Play Quasi-Newton Methodscited by 15
  2. 2022On Maximum-a-Posteriori estimation with Plug & Play priors and stochastic gradient descentcited by 14
  3. 2023Convergent Bregman Plug-and-Play Image Restoration for Poisson Inverse Problemscited by 14
  4. 2023A relaxed proximal gradient descent algorithm for convergent plug-and-play with proximal denoisercited by 12
  5. 2023Prior Mismatch and Adaptation in PnP-ADMM with a Nonconvex Convergence Analysiscited by 10
  6. 2022Bayesian Inversion for Nonlinear Imaging Models using Deep Generative Priorscited by 9
  7. 2022Deep Model-Based Architectures for Inverse Problems under Mismatched Priorscited by 8
  8. 2023Block Coordinate Plug-and-Play Methods for Blind Inverse Problemscited by 8
  9. 2024Patch-Based Diffusion Models Beat Whole-Image Models for Mismatched Distribution Inverse Problemscited by 8
  10. 2025Proximal Algorithm Unrolling: Flexible and Efficient Reconstruction Networks for Single-Pixel Imagingcited by 8
  11. 2022Memory-efficient model-based deep learning with convergence and robustness guaranteescited by 7
  12. 2025Deep Equilibrium models for Poisson Imaging Inverse problems via Mirror Descentcited by 7
  13. 2022A Plug-and-Play Approach to Multiparametric Quantitative MRI: Image Reconstruction using Pre-Trained Deep Denoiserscited by 5
  14. 2022CoRRECT: A Deep Unfolding Framework for Motion-Corrected Quantitative R2* Mappingcited by 5
  15. 2023Convergent plug-and-play with proximal denoiser and unconstrained regularization parametercited by 5
  16. 2024Plug-and-Play Priors as a Score-Based Methodcited by 5
  17. 2025MAP Estimation with Denoisers: Convergence Rates and Guaranteescited by 5
  18. 2021Unrolled Variational Bayesian Algorithm for Image Blind Deconvolutioncited by 4
  19. 2022Bregman Plug-and-Play Priorscited by 3
  20. 2023Convergence of Nonconvex PnP-ADMM with MMSE Denoiserscited by 3
  21. 2025PSI3D: Plug-and-Play 3D Stochastic Inference with Slice-wise Latent Diffusion Priorcited by 3
  22. 2026Stochastic Generative Plug-and-Play Priorscited by 3
  23. 2022Accelerating Plug-and-Play Image Reconstruction via Multi-Stage Sketched Gradientscited by 2
  24. 2024Parseval Convolution Operators and Neural Networkscited by 2
  25. 2023Local monotone operator learning using non-monotone operators: MnM-MOLcited by 1
  26. 2025Universal Architectures for the Learning of Polyhedral Norms and Convex Regularizerscited by 1
  27. 2026ShuffleFlow: Scalable Posterior Inference for Bayesian Inverse Imagingcited by 1
  28. 2022To what extent can Plug-and-Play methods outperform neural networks alone in low-dose CT reconstructioncited by 0
  29. 2022Operator Sketching for Deep Unrolling Networkscited by 0
  30. 2024LAMA: Stable Dual-Domain Deep Reconstruction For Sparse-View CTcited by 0
  31. 2025Enhanced Denoising and Convergent Regularisation Using Tweedie Scalingcited by 0
  32. 2025LAMA-Net: A Convergent Network Architecture for Dual-Domain Reconstructioncited by 0
  33. 2025Designing across domains with declarative thinking: Insights from the 96-Eyes ptychographic imager projectcited by 0
  34. 2025From the Gradient-Step Denoiser to the Proximal Denoiser and their associated convergent Plug-and-Play algorithmscited by 0
  35. 2025Beyond sparse denoising in frames: minimax estimation with a scattering transformcited by 0
  36. 2026Convergent Plug-and-Play Image Restoration with Annealed Noise Levelscited by 0
  37. 2026Plug-and-Play Methods Provably Converge Even with Improperly Trained Denoisers: Convergence by Architectural Designcited by 0