Research map / Time series / Causal and generative time series
Learning linear state-space models with sparse system matrices
Yasen Wang, Kaiqi Fang, Guijun Ma, Junlin Li, Mengyu Sun, Zhilan Huang, Gang Lu
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 |
|---|---|---|---|
| add Student-t sparsity prior with EM MAP estimation | MLE (EM) for LSSM · Evaporation system | MRE: 20.14% -> 14.93% | Improved |
| add sparsity prior | MLE · Glass furnace | MRE: 30.21% -> 23.63% | Improved |
| add sparsity prior | MLE · Steam generator | MRE: 22.23% -> 20.70% | Improved |
| add sparsity prior | MLE · Synthetic 10-dim system | average MRE: 5.97% -> 5.67% | No clear effect |
| add sparsity prior | MLE / classical algorithms · Synthetic 10-dim system | topology recovery success rate: 0 -> 95% | Improved |
| vary initialization a from 0.6 to 1.4 (ablation) | Sparse LSSM EM · Synthetic 10-dim system | MRE: 5.70% -> 5.69% | No clear effect |
| use sparse EM instead of 4SID | 4SID · Steam generator | runtime (s): 0.58 -> 441.88 | Hurt |
Closest papers by meaning
- Probabilistic Recurrent State-Space Models (ICML 2018 · similarity 0.88)
- Learning Low-dimensional Latent Dynamics from High-dimensional Observations: Non-asymptotics and Lower Bounds (ICML 2024 · similarity 0.86)
- The Curious Case of In-Training Compression of State Space Models (ICLR 2026 · similarity 0.86)
- Structured Sparse Transition Matrices to Enable State Tracking in State-Space Models (NeurIPS 2025 · similarity 0.86)
- HOPE for a Robust Parameterization of Long-memory State Space Models (ICLR 2025 · similarity 0.85)
- From Generalization Analysis to Optimization Designs for State Space Models (ICML 2024 · 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.
