Summary of Model Order Reduction Of Deep Structured State-space Models: a System-theoretic Approach, by Marco Forgione et al.
Model order reduction of deep structured state-space models: A system-theoretic approach
by Marco Forgione, Manas Mejari, Dario Piga
First submitted to arxiv on: 21 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Systems and Control (eess.SY)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper addresses the challenge of achieving accurate system modeling with limited complexity in parametric system identification, particularly for control design objectives. It focuses on deep structured state-space models (SSSM) that feature linear dynamical blocks as key components, offering high predictive performance but often suffering from excessively large model orders. To address this issue, the authors introduce two regularization terms: modal _1 and Hankel nuclear norm regularization, which can be incorporated into the training loss for improved model order reduction. These regularizers promote sparsity, allowing retention of only relevant states without sacrificing accuracy. The methodology is demonstrated using real-world ground vibration data from an aircraft. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper aims to improve system modeling in parametric system identification by reducing the complexity of deep structured state-space models (SSSM). This is done by introducing two new regularization techniques that help retain only important information, making it easier to use these models for control design. The results show that this approach can create more accurate and faster models. |
Keywords
* Artificial intelligence * Regularization
