Summary of Learnable & Interpretable Model Combination in Dynamical Systems Modeling, by Tobias Thummerer and Lars Mikelsons
Learnable & Interpretable Model Combination in Dynamical Systems Modeling
by Tobias Thummerer, Lars Mikelsons
First submitted to arxiv on: 12 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 The proposed work discusses combining multiple model architectures for dynamical systems modeling, including machine learning and physical simulation models. The authors propose a new class of models that can express mixed algebraic, discrete, and differential equation-based models. They also examine different ways to combine these models from the perspective of system theory and highlight two challenges: algebraic loops and local event functions in discontinuous models. To address these challenges, they introduce a wildcard architecture that can describe arbitrary combinations of models and can be learned through gradient-based optimization. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper explores how to combine different model architectures for dynamical systems modeling. They propose a new type of model that can mix together algebraic, discrete, and differential equation-based models. The authors also look at different ways to combine these models from the perspective of system theory. Two challenges they identify are algebraic loops and local event functions in discontinuous models. To solve these problems, they introduce a special type of architecture that can describe any combination of models. They show how this architecture can be learned using gradient-based optimization. |
Keywords
» Artificial intelligence » Machine learning » Optimization
