Discovering governing equations from partial measurements with deep delay autoencoders
Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences · 2023-08 · 479 (2276)
DOI: 10.1098/rspa.2023.0422 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, online erschienen am 30.08.2023. Verfasst von Joseph Bakarji und 3 weiteren Autor:innen. Die Arbeit wurde bisher 88-mal zitiert (Crossref) und nennt 70 Literaturangaben. Gefördert von: National Science Foundation AI Institute in Dynamic Systems, Army Research Office.
Abstract (Originalsprache)
A central challenge in data-driven model discovery is the presence of hidden, or latent, variables that are not directly measured but are dynamically important. Takens’ theorem provides conditions for when it is possible to augment partial measurements with time delayed information, resulting in an attractor that is diffeomorphic to that of the original full-state system. This diffeomorphism is typically unknown, and learning the dynamics in the embedding space has remained an open challenge for decades. Here, we design a deep autoencoder network to learn a coordinate transformation from the delay embedded space into a new space, where it is possible to represent the dynamics in a sparse, closed form. We demonstrate this approach on the Lorenz, Rössler and Lotka–Volterra systems, as well as a Lorenz analogue from a video of a chaotic waterwheel experiment. This framework combines deep learning and the sparse identification of nonlinear dynamics methods to uncover interpretable models within effective coordinates.
Förderung
- National Science Foundation AI Institute in Dynamic Systems
- Army Research Office
Zitieren
APA
Bakarji, J., Champion, K., Kutz, J. N., & Brunton, S. L. (2023). Discovering governing equations from partial measurements with deep delay autoencoders. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences, 479(2276). https://doi.org/10.1098/rspa.2023.0422
Vancouver
Bakarji J, Champion K, Kutz JN, Brunton SL. Discovering governing equations from partial measurements with deep delay autoencoders. Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences. 2023;479(2276). doi:10.1098/rspa.2023.0422
BibTeX
@article{Bakarji2023,
author = {Bakarji, Joseph and Champion, Kathleen and Kutz, J. Nathan and Brunton, Steven L.},
title = {Discovering governing equations from partial measurements with deep delay autoencoders},
journal = {Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences},
year = {2023},
volume = {479},
number = {2276},
doi = {10.1098/rspa.2023.0422}
}
EUVIMED