Variational Data Assimilation and Model Learning

Assimilation variationnelle de donnees et apprentissage modeles

Jerome Monnier (1, 2)

1 INSA - Institut National des Sciences Appliquées
2 IMT - Institut de Mathématiques de Toulouse UMR5219

To assimilate datasets into physically-based models, to efficiently compute a model-output gradient, to calibrate models / identify model parameters, to learn model terms by Jérôme Monnier Fall 2021

Mots clés: inverse problems; model-based feedback control; PDE models; optimisation; gradient-based algorithms; adjoint method; best estimate; error covariances; Earth sciences; model learning; inverse problems; gradientbased algorithms

Domaines: Analyse numérique [math.NA]; Mathématiques [math]; Optimisation et contrôle [math.OC]; Sciences de l'environnement; Sciences de la Terre; Sciences de l'ingénieur [physics]

Jerome Monnier. Variational Data Assimilation and Model Learning. Master. France. 2021. ⟨hal-03040047v2⟩