Freezing pretrained variable-specific experts and adding new experts for surface variables lets a weather transformer match full retraining accuracy with about a quarter of the training iterations.
Accurate medium-range global weather forecasting with 3d neural networks.Nature, 619(7970): 533–538, 2023
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
VA-MoE: Variables-Adaptive Mixture of Experts for Incremental Weather Forecasting
Freezing pretrained variable-specific experts and adding new experts for surface variables lets a weather transformer match full retraining accuracy with about a quarter of the training iterations.