Learning 4DVAR inversion directly from observations - Sorbonne Université
Conference Papers Year : 2023

Learning 4DVAR inversion directly from observations

Abstract

Variational data assimilation and deep learning share many algorithmic aspects in common. While the former focuses on system state estimation, the latter provides great inductive biases to learn complex relationships. We here design a hybrid architecture learning the assimilation task directly from partial and noisy observations, using the mechanistic constraint of the 4DVAR algorithm. Finally, we show in an experiment that the proposed method was able to learn the desired inversion with interesting regularizing properties and that it also has computational interests.
Fichier principal
Vignette du fichier
ICCS_MLDADS2023.pdf (661.04 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-03863390 , version 1 (21-11-2022)
hal-03863390 , version 2 (05-04-2023)

Identifiers

  • HAL Id : hal-03863390 , version 2

Cite

Arthur Filoche, Julien Brajard, Anastase Alexandre Charantonis, Dominique Béréziat. Learning 4DVAR inversion directly from observations. Workshop on Machine Learning and Data Assimilation for Dynamical Systems (MLDADS), International Conference on Computational Science (ICCS), Jul 2023, Prague, Czech Republic. ⟨hal-03863390v2⟩
264 View
208 Download

Share

More