mlsanityData to 7 Oct 2026

Optimization for Amortized Inverse Problems

arXiv 2210.13983 ↗First posted 25 Oct 2022cs.LGcited by 6

Not on the map. Why?

Abstract (arXiv)

Incorporating a deep generative model as the prior distribution in inverse problems has established substantial success in reconstructing images from corrupted observations. Notwithstanding, the existing optimization approaches use gradient descent largely without adapting to the non-convex nature of the problem and can be sensitive to initial values, impeding further performance improvement. In this paper, we propose an efficient amortized optimization scheme for inverse problems with a deep generative prior. Specifically, the optimization task with high degrees of difficulty is decomposed into optimizing a sequence of much easier ones. We provide a theoretical guarantee of the proposed algorithm and empirically validate it on different inverse problems. As a result, our approach outperforms baseline methods qualitatively and quantitatively by a large margin.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 45 of its 65 references that are in the map. First posted on arXiv 25 Oct 2022.

Builds on

45 of 65 references.

References · 45

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  3. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  4. 2013Auto-Encoding Variational Bayescited by 3,980
  5. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  6. 2017Progressive Growing of GANs for Improved Quality, Stability, and Variationcited by 2,967
  7. 2018Neural Tangent Kernel: Convergence and Generalization in Neural Networkscited by 2,597
  8. 2021Convex optimizationcited by 2,479
  9. 2015Variational Inference with Normalizing Flowscited by 2,161
  10. 2016Density estimation using Real NVPcited by 2,029
  11. 2018Glow: Generative Flow with Invertible 1x1 Convolutionscited by 1,539
  12. 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,315
33 more
  1. 2017Deep Image Priorcited by 1,097
  2. 2018A Convergence Theory for Deep Learning via Over-Parameterizationcited by 784
  3. 2018Gradient Descent Finds Global Minima of Deep Neural Networkscited by 572
  4. 2016Linear Convergence of Gradient and Proximal-Gradient Methods Under the Polyak-Lojasiewicz Conditioncited by 526
  5. 2020PULSE: Self-Supervised Photo Upsampling via Latent Space Exploration of Generative Modelscited by 358
  6. 2017Compressed Sensing using Generative Modelscited by 337
  7. 2020Deep Learning Techniques for Inverse Problems in Imagingcited by 290
  8. 2021Robust Compressed Sensing MRI with Deep Generative Priorscited by 284
  9. 2020Loss landscapes and optimization in over-parameterized non-linear systems and neural networkscited by 231
  10. 2021Learning to Optimize: A Primer and A Benchmarkcited by 206
  11. 2017Convergence Analysis of Two-layer Neural Networks with ReLU Activationcited by 137
  12. 2018Fast and Scalable Bayesian Deep Learning by Weight-Perturbation in Adamcited by 131
  13. 2018Iterative Amortized Inferencecited by 93
  14. 2019Invertible generative models for inverse problems: mitigating representation error and dataset biascited by 93
  15. 2019Graduated Non-Convexity for Robust Spatial Perception: From Non-Minimal Solvers to Global Outlier Rejectioncited by 92
  16. 2018Semi-Amortized Variational Autoencoderscited by 83
  17. 2018Phase Retrieval Under a Generative Priorcited by 73
  18. 2017Natasha 2: Faster Non-Convex Optimization Than SGDcited by 49
  19. 2020Composing Normalizing Flows for Inverse Problemscited by 48
  20. 2018Compressed Sensing with Deep Image Prior and Learned Regularizationcited by 47
  21. 2018Blind Image Deconvolution using Deep Generative Priorscited by 40
  22. 2019SGD Converges to Global Minimum in Deep Learning via Star-convex Pathcited by 39
  23. 2017Global Guarantees for Enforcing Deep Generative Priors by Empirical Riskcited by 37
  24. 2015On Graduated Optimization for Stochastic Non-Convex Problemscited by 34
  25. 2019Inverting Deep Generative models, One layer at a timecited by 33
  26. 2020The Effects of Mild Over-parameterization on the Optimization Landscape of Shallow ReLU Neural Networkscited by 33
  27. 2020Robust Compressed Sensing using Generative Modelscited by 33
  28. 2020Solving Inverse Problems with a Flow-based Noise Modelcited by 31
  29. 2021Accelerating Quadratic Optimization with Reinforcement Learningcited by 30
  30. 2021Subquadratic Overparameterization for Shallow Neural Networkscited by 26
  31. 2022Teaching Networks to Solve Optimization Problemscited by 7
  32. 2019Surfing: Iterative optimization over incrementally trained deep networkscited by 5
  33. 2018Semantic Image Inpainting Through Improved Wasserstein Generative Adversarial Networkscited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 1; 2024: 2; 2025: 2; 2026: 1.

6 papers.

  1. 2023Deep Injective Prior for Inverse Scatteringcited by 3
  2. 2024Back-Projection Diffusion: Solving the Wideband Inverse Scattering Problem with Diffusion Modelscited by 3
  3. 2024LoFi: Neural Local Fields for Scalable Image Reconstructioncited by 2
  4. 2025Rare event modeling with self-regularized normalizing flows: what can we learn from a single failure?cited by 1
  5. 2026Tunable Latent Generative Priors for Compressed Sensing and Inverse Problemscited by 1
  6. 2025Beyond Invisibility: Learning Robust Visible Watermarks for Stronger Copyright Protectioncited by 0