Research map / Learning theory and algorithms / Matrix and tensor methods
Finite Time LTI System Identification
Tuhin Sarkar, Alexander Rakhlin, Munther A. Dahleh
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 the proposed data-dependent Hankel order selection instead of SSREGEST (large model then reduction) | Hankel OLS system identification · Synthetic order-150 LTI system, rho=0.9, T=2000 | model error (Hankel norm): 26.38 -> 7.70 | Improved |
| use the proposed method instead of SSREGEST at very short data length | Hankel OLS system identification · Synthetic order-150 LTI system, rho=0.9, T=500 | model error (Hankel norm): 6.21 -> 13.37 | Hurt |
| use adaptive Hankel size d-hat instead of deterministic growth d = log(T) (ablation) | Hankel OLS system identification · Synthetic LTI, rho=0.99 | model error (Hankel norm) | Improved |
| use adaptive Hankel size d-hat instead of deterministic growth d = log(T) (ablation) | Hankel OLS system identification · Synthetic LTI, rho=0.6 | model error (Hankel norm) | No clear effect |
| increase spectral radius rho from 0.6 to 0.99 | selected Hankel size d-hat · Synthetic LTI, varying T | selected Hankel size | Mixed |
Closest papers by meaning
- Near optimal finite time identification of arbitrary linear dynamical systems (ICML 2019 · similarity 0.90)
- Stabilizing LTI Systems under Partial Observability: Sample Complexity and Fundamental Limits (NeurIPS 2025 · similarity 0.87)
- Learning Low-dimensional Latent Dynamics from High-dimensional Observations: Non-asymptotics and Lower Bounds (ICML 2024 · similarity 0.85)
- Finite Sample Analyses for Continuous-time Linear Systems: System Identification and Online Control (NeurIPS 2025 · similarity 0.85)
- HOPE for a Robust Parameterization of Long-memory State Space Models (ICLR 2025 · similarity 0.85)
- PAC-Bayesian Error Bound, via Rényi Divergence, for a Class of Linear Time-Invariant State-Space Models (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.
