Summary of Neural Ddes with Learnable Delays For Partially Observed Dynamical Systems, by Thibault Monsel et al.
Neural DDEs with Learnable Delays for Partially Observed Dynamical Systems
by Thibault Monsel, Emmanuel Menier, Onofrio Semeraro, Lionel Mathelin, Guillaume Charpiat
First submitted to arxiv on: 3 Oct 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Computational Physics (physics.comp-ph)
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 introduces a new approach to learning dynamical systems from data, building upon the Mori-Zwanzig formalism from statistical physics. The proposed Constant Lag Neural Delay Differential Equations (NDDEs) naturally model partially observed states, which is a common scenario in practice. In experiments, the authors demonstrate that NDDEs outperform existing methods on both synthetic and real-world datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps us learn more about complex systems from data. Right now, many ways to do this rely on having all the information we need. But often, we don’t have that luxury – we only see some parts of the system. The authors suggest a new way to model these partial observations using something called Constant Lag Neural Delay Differential Equations (NDDEs). They show that NDDEs work better than other methods on both made-up and real-world data. |
