Skip to main content

Ctrl+K

Learning Machine Learning with Lorenz-96

Lorenz-96 and General Circulation Models

The Lorenz-96 Two-Timescale System
The Lorenz-96 and its GCM Analog
GCM parameterizations, skill metrics, and other sources of uncertainity
Tuning GCM Parameterizations

Neural Networks with Lorenz-96

Introduction to Machine Learning and Neural Networks
Using Neural Networks for L96 Parameterization: Offline Training
Using Neural Networks for L96 Parameterization: Online Testing
Using Neural Networks for L96 Parameterization: Online Training
Improving Performance of Neural Networks
Interpreting Neural Networks
Adding constraints to Neural Networks

Data Assimilation with Lorenz-96

Data Assimilation demo in the Lorenz 96 (L96) two time-scale model
Learning Data Assimilation Increments

Equation Discovery with Lorenz-96

Introduction to Equation Discovery - Comparing Symbolic Regression Methods
Applying SINDy equation identification to L96
Symbolic Regression vs. Neural Networks on Multiscale L96

Other ML approaches for Lorenz-96

Random Forest
Gaussian Processes
CNNs and Polynomial Maps
ResNet and Recurrent Neural Networks

End Matter

Outlook
Citing this book
References

Index

By The M²LInES Community

© Copyright 2024 Dhruv Balwada, Ryan Abernathey, Shantanu Acharya, et al. — License: MIT for code, CC-BY for text and figures.