Research map / Optimization / Optimizers
Efficient Optimization for Linear Dynamical Systems with Applications to Clustering and 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 stochastic mini-batch PGD (batch 128) instead of full-batch PGD for dictionary learning (ablation) | PGD sparse coding · DynTex++ testing normalized reconstruction error | normalized reconstruction error | Helped |
| Replace Grass-3 and LDSST with PGD-128 for sparse coding | LDS dictionary learning · DynTex++ | learning speed and final error: more than 50 times faster per epoch | Helped |
| Grass-3 versus PGD-128 at small dictionary size J=4 | sparse coding · DynTex++ J=4 | final testing NR | Hurt |
| Replace MDS/Align clustering with PGD | LDS clustering · DynTex++ purity, 4-128 clusters | purity | Helped |
| Replace MDS/Align clustering with PGD | LDS clustering · DynTex++ | time per epoch | Helped |
| Use STPM pooling with sparse coding instead of BoS with clustering | PGD · YUPENN | accuracy (%): 84.1 -> 93.6 | Helped |
| Use STPM pooling with sparse coding instead of BoS with clustering | PGD · DynTex | accuracy (%): 65.4 -> 76.5 | Helped |
| Use PGD instead of Align in BoS | BoS · DynTex | accuracy (%): 62.7 -> 65.4 | Helped |
| Use PGD instead of Grass-3 in STPM | STPM · YUPENN | accuracy (%): 91.6 -> 93.6 | Helped |
| Use PGD instead of Grass-3 in STPM | STPM · DynTex | accuracy (%): 75.1 -> 76.5 | Helped |
Closest papers by meaning
- Sparse Coding and Dictionary Learning With Linear Dynamical Systems (CVPR 2016 · similarity 0.89)
- Memory-Efficient Learning of Stable Linear Dynamical Systems for Prediction and Control (NeurIPS 2020 · similarity 0.86)
- Learning Linear Dynamical Systems via Spectral Filtering (NeurIPS 2017 · similarity 0.86)
- Sparse Learning of Dynamical Systems in RKHS: An Operator-Theoretic Approach (ICML 2023 · similarity 0.86)
- Sparse Diffusion Autoencoder for Test-time Adapting Prediction of Complex Systems (NeurIPS 2025 · similarity 0.84)
- Approximate Gradient Coding for Distributed Learning with Heterogeneous Stragglers (NeurIPS 2025 · 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.