paper
2023
Hybrid physics-ML & parameterization
New model or method
RNN / LSTM / ConvLSTMDifferentiable physical modelColumbia University (LEAP, Lamont-Doherty) Journal
2023-07-01
022
Jul 2023
Mohamed Aziz Bhouri, Pierre Gentine
Chaos An Interdisciplinary Journal of Nonlinear Science
Columbia University (LEAP, Lamont-Doherty)
Proposes memory-based neural network closures trained with a differentiable solver for Lorenz '96, giving more stable long-horizon forecasts than instantaneous parameterizations.
Hybrid physics-ML & parameterization
Memory-based parameterization with differentiable solver: Application to Lorenz ’96
In one sentence. Proposes memory-based neural network closures trained with a differentiable solver for Lorenz '96, giving more stable long-horizon forecasts than instantaneous parameterizations.
Labels
- Topic
- Hybrid physics-ML & parameterization
- Type
- New model or method
- Method
- RNN / LSTM / ConvLSTM, Differentiable physical model
- Domain
- Global
- Data
- Other
- Status
- Research
- Identifiers
- DOI:10.1063/5.0131929 OpenAlex:W4383302406
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.