mlsanityData to 7 Oct 2026

WARPd: A linearly convergent first-order method for inverse problems with approximate sharpness conditions

arXiv 2110.12437 ↗First posted 24 Oct 2021math.NA cs.CV cs.IRcited by 1

Not on the map. Why?

Abstract (arXiv)

Reconstruction of signals from undersampled and noisy measurements is a topic of considerable interest. Sharpness conditions directly control the recovery performance of restart schemes for first-order methods without the need for restrictive assumptions such as strong convexity. However, they are challenging to apply in the presence of noise or approximate model classes (e.g., approximate sparsity). We provide a first-order method: Weighted, Accelerated and Restarted Primal-dual (WARPd), based on primal-dual iterations and a novel restart-reweight scheme. Under a generic approximate sharpness condition, WARPd achieves stable linear convergence to the desired vector. Many problems of interest fit into this framework. For example, we analyze sparse recovery in compressed sensing, low-rank matrix recovery, matrix completion, TV regularization, minimization of |Bx|l1 under constraints (l1-analysis problems for general B), and mixed regularization problems. We show how several quantities controlling recovery performance also provide explicit approximate sharpness constants. Numerical experiments show that WARPd compares favorably with specialized state-of-the-art methods and is ideally suited for solving large-scale problems. We also present a noise-blind variant based on the Square-Root LASSO decoder. Finally, we show how to unroll WARPd as neural networks. This approximation theory result provides lower bounds for stable and accurate neural networks for inverse problems and sheds light on architecture choices. Code and a gallery of examples are made available online as a MATLAB package.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 25 of its 148 references that are in the map. First posted on arXiv 24 Oct 2021.

Builds on

25 of 148 references.

References · 25

  1. 2016Deep Convolutional Neural Network for Inverse Problems in Imagingcited by 431
  2. 2019Algorithm Unrolling: Interpretable, Efficient Deep Learning for Signal and Image Processingcited by 411
  3. 2017Learning a Variational Network for Reconstruction of Accelerated MRI Datacited by 340
  4. 2017Training GANs with Optimismcited by 217
  5. 2019On instabilities of deep learning in image reconstruction - Does AI come at a cost?cited by 211
  6. 2018Theoretical Linear Convergence of Unfolded ISTA and its Practical Weights and Thresholdscited by 83
  7. 2017A Review of Convolutional Neural Networks for Inverse Problems in Imagingcited by 81
  8. 2020Advancing machine learning for MR image reconstruction with an open competition: Overview of the 2019 fastMRI challengecited by 59
  9. 2020The gap between theory and practice in function approximation with deep neural networkscited by 53
  10. 2021Can stable and accurate neural networks be computed? -- On the barriers of deep learning and Smale's 18th problemcited by 53
  11. 2015From error bounds to the complexity of first-order descent methods for convex functionscited by 42
  12. 2017Stochastic Primal-Dual Hybrid Gradient Algorithm with Arbitrary Sampling and Imaging Applicationscited by 29
13 more
  1. 2018Learning The Invisible: A Hybrid Deep Learning-Shearlet Framework for Limited Angle Computed Tomographycited by 29
  2. 2014Splitting methods with variable metric for KL functionscited by 26
  3. 2015A Unified Approach to Error Bounds for Structured Convex Optimization Problemscited by 21
  4. 2020The Numerics of Phase Retrievalcited by 20
  5. 2020Results of the 2020 fastMRI Challenge for Machine Learning MR Image Reconstructioncited by 14
  6. 2014Local Linear Convergence of Forward-Backward under Partial Smoothnesscited by 13
  7. 2021Faster First-Order Primal-Dual Methods for Linear Programming using Restarts and Sharpnesscited by 13
  8. 2020Total Deep Variation: A Stable Regularizer for Inverse Problemscited by 9
  9. 2015New Computational Guarantees for Solving Convex Optimization Problems with First Order Methods, via a Function Growth Condition Measurecited by 5
  10. 2019On the convex geometry of blind deconvolution and matrix completioncited by 4
  11. 2016Restarting accelerated gradient methods with a rough strong convexity estimatecited by 3
  12. 2021Proof methods for robust low-rank matrix recoverycited by 3
  13. 2016A multilevel based reweighting algorithm with joint regularizers for sparse recoverycited by 1

Built on by

Cited by year

Citing papers per year: 2022: 1.

1 paper.

  1. 2022On efficient algorithms for computing near-best polynomial approximations to high-dimensional, Hilbert-valued functions from limited samplescited by 8