Research map / Learning theory and algorithms / Matrix and tensor methods
SC2Net: Sparse LSTMs 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 |
|---|---|---|---|
| replace simple RNN unit of LISTA with SC2Net LSTM-based architecture (ablation) | LISTA · MNIST unsupervised | accuracy (%): 85.25 -> 88.75 | Helped |
| replace simple RNN unit of LISTA with SC2Net architecture (ablation) | LISTA · CIFAR-10 unsupervised | accuracy (%): 42.12 -> 60.20 | Helped |
| replace simple RNN unit of LISTA with SC2Net architecture (ablation) | LISTA · CIFAR-10 supervised | accuracy (%): 52.35 -> 71.10 | Helped |
| replace simple RNN unit of LFISTA with SC2Net architecture (ablation) | LFISTA · MNIST supervised | accuracy (%): 87.95 -> 91.60 | Helped |
| replace simple RNN unit of LFISTA with SC2Net architecture (ablation) | LFISTA · CIFAR-10 supervised | accuracy (%): 53.10 -> 72.48 | Helped |
| use SLSTM instead of LFISTA | sparse coding solver · CIFAR-10 unsupervised | accuracy (%): 42.75 -> 65.40 | Helped |
| use SLSTM instead of LFISTA | sparse coding solver · MNIST supervised | accuracy (%): 87.95 -> 94.31 | Helped |
| use SLSTM instead of LISTA | sparse coding solver · CIFAR-10 supervised | accuracy (%): 52.35 -> 79.53 | Helped |
| add label supervision to SLSTM | SLSTM · CIFAR-10 | accuracy (%): 65.40 -> 79.53 | Helped |
Closest papers by meaning
- Learning step sizes for unfolded sparse coding (NeurIPS 2019 · similarity 0.90)
- Sparse Coding with Gated Learned ISTA (ICLR 2020 · similarity 0.89)
- A Single-Step, Sharpness-Aware Minimization is All You Need to Achieve Efficient and Accurate Sparse Training (NeurIPS 2024 · similarity 0.85)
- Global Minimizers of $\ell^p$-Regularized Objectives Yield the Sparsest ReLU Neural Networks (NeurIPS 2025 · similarity 0.85)
- Two Sparse Matrices are Better than One: Sparsifying Neural Networks with Double Sparse Factorization (ICLR 2025 · similarity 0.85)
- Sparse-IFT: Sparse Iso-FLOP Transformations for Maximizing Training Efficiency (ICML 2024 · 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.