Neural Koopman prior for data assimilation - IMT Atlantique Access content directly
Journal Articles IEEE Transactions on Signal Processing Year : 2024

Neural Koopman prior for data assimilation


With the increasing availability of large scale datasets, computational power and tools like automatic differentiation and expressive neural network architectures, sequential data are now often treated in a data-driven way, with a dynamical model trained from the observation data. While neural networks are often seen as uninterpretable black-box architectures, they can still benefit from physical priors on the data and from mathematical knowledge. In this paper, we use a neural network architecture which leverages the long-known Koopman operator theory to embed dynamical systems in latent spaces where their dynamics can be described linearly, enabling a number of appealing features. We introduce methods that enable to train such a model for long-term continuous reconstruction, even in difficult contexts where the data comes in irregularly-sampled time series. The potential for self-supervised learning is also demonstrated, as we show the promising use of trained dynamical models as priors for variational data assimilation techniques, with applications to e.g. time series interpolation and forecasting.
Fichier principal
Vignette du fichier
bare_jrnl_new_sample4.pdf (2.8 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04182289 , version 1 (17-08-2023)
hal-04182289 , version 2 (06-03-2024)
hal-04182289 , version 3 (19-06-2024)




Anthony Frion, Lucas Drumetz, Mauro Dalla Mura, Guillaume Tochon, Abdeldjalil Aissa El Bey. Neural Koopman prior for data assimilation. IEEE Transactions on Signal Processing, inPress. ⟨hal-04182289v3⟩
143 View
198 Download



Gmail Mastodon Facebook X LinkedIn More