Reformulates ML weather model rollout training as 4DVar in autoencoder latent space to approximate cross-variable error covariances and improve physical realism in long-term forecasts.
Prediction error growth in a more realistic atmospheric toy model with three spatiotemporal scales.Geoscientific Model Development, 15(10):4147–4161, May 2022
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Learning more physically realistic dynamics in machine-learning based weather forecasting with latent-space constraints
Reformulates ML weather model rollout training as 4DVar in autoencoder latent space to approximate cross-variable error covariances and improve physical realism in long-term forecasts.