Research map / Learning theory and algorithms / Matrix and tensor methods
Learned Extragradient ISTA with Interpretable Residual Structures for Sparse Coding
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 |
|---|---|---|---|
| Use ELISTA rather than LFISTA | Natural image inpainting · Barbara, Boat, House, Lena, Peppers, C.man, Couple, Finger, Man, Montage | PSNR: Barbara: 26.01 -> 26.60; Boat: 29.68 -> 30.33; House: 32.06 -> 32.76; Lena: 32.12 -> 32.75; Peppers: 28.57 -> 29.61; C.m | Helped |
| Use ELISTA rather than LFISTA | Natural image inpainting · Hill | PSNR: 30.69 -> 30.41 | Hurt |
| Use ELISTA rather than GLISTA | Photometric stereo · q=35 | mean angular error: 0.06249 -> 0.02754 | Helped |
| Use ELISTA rather than GLISTA | Photometric stereo · q=25 | mean angular error: 0.10033 -> 0.04947 | Helped |
| Use ELISTA rather than GLISTA | Photometric stereo · q=15 | mean angular error: 0.63967 -> 0.60010 | Helped |
| Add the extragradient residual structure (ablation) | LISTA-m · Sparse representation, noiseless test set | NMSE (dB): −63.73 -> −107.48 | Helped |
| Add the extragradient residual structure (ablation) | LISTA-m · Sparse representation, noiseless test set | SPERR: 57.22 -> 0.01 | Helped |
| Use MT instead of soft thresholding (ablation) | ELISTA-m · Sparse representation, noiseless test set | NMSE (dB): −63.73 -> −107.48 | Helped |
| Use MT instead of soft thresholding (ablation) | ELISTA-m · Sparse representation, noiseless test set | SPERR: 57.22 -> 0.01 | Helped |
| Tie the algorithm parameters (ablation) | ELISTA-t · Sparse representation, noiseless test set | NMSE (dB): −77.03 -> −107.48 | Helped |
| Tie the algorithm parameters (ablation) | ELISTA-t · Sparse representation, noiseless test set | SPERR: 0.15 -> 0.01 | Helped |
Closest papers by meaning
- Sparse Coding with Gated Learned ISTA (ICLR 2020 · similarity 0.94)
- Learning step sizes for unfolded sparse coding (NeurIPS 2019 · similarity 0.90)
- Theoretical Linear Convergence of Unfolded ISTA and Its Practical Weights and Thresholds (NeurIPS 2018 · similarity 0.88)
- SC2Net: Sparse LSTMs for Sparse Coding (AAAI 2018 · similarity 0.88)
- Understanding Trainable Sparse Coding with Matrix Factorization (ICLR 2017 · similarity 0.87)
- SequentialAttention++ for Block Sparsification: Differentiable Pruning Meets Combinatorial Optimization (NeurIPS 2024 · similarity 0.83)
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.