Woohoo our Neural General Circulation Model paper is out in Nature!! www.nature.com/articles/s41... How to combine a conventional simulation of the Earth's atmospheric circulation with learned physics (neural networks!) of atmospheric processes in clouds, radiation, surface fluxes, precipitation, ..
A hybrid model that combines a differentiable solver for atmospheric dynamics with machine-learning components is capable of weather forecasts and climate simulations on par with the best machine...
and train both together to get the best of worlds: Respecting physical laws and simulation of known dynamics conventionally, but correcting this dynamical core towards data with neural networks. A climate model that can learn from data. Sure, it's not perfect, for example