Research map / Optimization / Optimizers
A Flexible Framework for Designing Trainable Priors with Adaptive Smoothing and Game Encoding
Bruno Lecouat, Jean Ponce, Julien Mairal
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| Use asymmetric rather than symmetric TV weights (ablation) | Trainable TV · BSD68, σ=25 | average PSNR: 27.82 -> 28.24 | Improved |
| Use extra-gradient rather than gradient descent (ablation) | Trainable TV · BSD68, σ=25 | average PSNR: 27.58 -> 27.82 | Improved |
| Use extra-gradient rather than gradient descent (ablation) | Trainable TV asymmetric · BSD68, σ=25 | average PSNR: 27.99 -> 28.24 | Improved |
| Add variance regularization (ablation) | Sparse Coding · BSD68, σ=25 | average PSNR: 28.90 -> 29.00 | Improved |
| Add TV regularization (ablation) | Sparse Coding · BSD68, σ=25 | average PSNR: 28.90 -> 29.02 | Improved |
| Add variance regularization to sparse coding (ablation) | CS-Sparse coding · MR brain dataset, pseudo radial sampling, 30% | PSNR (dB): 40.50 -> 40.67 | Improved |
| Add variance regularization to sparse coding (ablation) | CS-Sparse coding · MR brain dataset, pseudo radial sampling, 20% | PSNR (dB): 37.80 -> 37.79 | No clear effect |
| Use Barzilai-Borwein step sizes (ablation) | Sparse Coding · BSD68, σ=25 | average PSNR: 28.57 -> 28.82 | Improved |
| Add spectral normalization (ablation) | Sparse Coding · BSD68, σ=25 | average PSNR: 28.57 -> 28.69 | Improved |
| Add the TV smoothing block | PSMNet · KITTI2015 validation | 3-px error (%): 2.14 ± 0.04 -> 2.10 ± 0.03 | Improved |
Closest papers by meaning
- Algorithmic Guarantees for Inverse Imaging with Untrained Network Priors (NeurIPS 2019 · similarity 0.87)
- Unrolled denoising networks provably learn to perform optimal Bayesian inference (NeurIPS 2024 · similarity 0.86)
- What's in a Prior? Learned Proximal Networks for Inverse Problems (ICLR 2024 · similarity 0.85)
- Solving Neural Min-Max Games: The Role of Architecture, Initialization & Dynamics (NeurIPS 2025 · similarity 0.85)
- Random Sparse Lifts: Construction, Analysis and Convergence of finite sparse networks (ICLR 2024 · similarity 0.85)
- Preconditioners for the Stochastic Training of Neural Fields (CVPR 2025 · similarity 0.85)
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.
