paper
2018
Hybrid physics-ML & parameterization
New model or method
Classical ML (trees, GP, linear, kernels)University of Washington Journal
2018-06-22
036
Jun 2018
N. D. Brenowitz, C. S. Bretherton
Geophysical Research Letters
University of Washington
Neural network unified physics parameterization trained on 4-km aquaplanet data gives stable prognostic single-column tests better than CAM's scheme.
Hybrid physics-ML & parameterization
Prognostic Validation of a Neural Network Unified Physics Parameterization
In one sentence. Neural network unified physics parameterization trained on 4-km aquaplanet data gives stable prognostic single-column tests better than CAM's scheme.
Labels
- Topic
- Hybrid physics-ML & parameterization
- Type
- New model or method
- Method
- Classical ML (trees, GP, linear, kernels)
- Domain
- Global
- Data
- km-scale or high-res simulation
- Status
- Research
- Identifiers
- DOI:10.1029/2018gl078510 OpenAlex:W2808400960
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.