no code implementations • 11 Oct 2023 • Theo Archambault, Arthur Filoche, Anastase Charantonis, Dominique Bereziat, Sylvie Thiria
We also introduce a new way to transfer the learning from simulation to observations: supervised pre-training on our OSSE followed by unsupervised fine-tuning on satellite data.
no code implementations • 17 Nov 2022 • Arthur Filoche, Julien Brajard, Anastase Charantonis, Dominique Béréziat
Variational data assimilation and deep learning share many algorithmic aspects in common.
1 code implementation • 9 Dec 2020 • Vincent Bouget, Dominique Béréziat, Julien Brajard, Anastase Charantonis, Arthur Filoche
The network was compared to a similar architecture trained only on radar data, to a basic persistence model and to an approach based on optical flow.