Research map / Probabilistic and scientific modeling / Variational and generative sampling
An inner-loop free solution to inverse problems using deep neural networks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace gradient-descent inner loop for the z update with the learned inversion network | Inf-ADMM-ADNN · Synthetic Laplace data, strided convolution | convergence speed vs wall-clock time | Helped |
| replace gradient-descent inner loop for the z update with the learned inversion network | Inf-ADMM-ADNN · Synthetic Laplace data, strided convolution | objective vs ADMM iteration | No clear effect |
| use learned inversion network instead of exact closed-form matrix inverse | Inf-ADMM-ADNN · Synthetic Laplace data | objective convergence | No clear effect |
| use Inf-ADMM-ADNN instead of bicubic interpolation | Super-resolution · MNIST and CUB-200-2011 | visual quality | Helped |
| use Inf-ADMM-ADNN instead of tuned Wiener filter | Motion deblurring · CUB-200-2011 | visual quality | Helped |
| use large ADMM beta (0.1 or 0.01) instead of smaller beta | Inf-ADMM-ADNN · MNIST super-resolution | NMSE convergence | Mixed |
| reduce ADMM beta to 0.0001, 0.0005 or 0.001 | Inf-ADMM-ADNN · MNIST super-resolution | NMSE | Mixed |
| run more ADMM iterations (1st, 4th, 7th) | Inf-ADMM-ADNN · CelebA joint super-resolution and colorization | visual quality | Helped |
Closest papers by meaning
- NPN: Non-Linear Projections of the Null-Space for Imaging Inverse Problems (NeurIPS 2025 · similarity 0.85)
- Integrating Intermediate Layer Optimization and Projected Gradient Descent for Solving Inverse Problems with Diffusion Models (ICML 2025 · similarity 0.85)
- Self-diffusion for Solving Inverse Problems (NeurIPS 2025 · similarity 0.85)
- What's in a Prior? Learned Proximal Networks for Inverse Problems (ICLR 2024 · similarity 0.85)
- A Restoration Network as an Implicit Prior (ICLR 2024 · similarity 0.85)
- Improving Diffusion-based Inverse Algorithms under Few-Step Constraint via Linear Extrapolation (NeurIPS 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.