# Coupled online learning as a way to tackle instabilities and biases in neural network parameterizations

## TL;DR

This paper introduces coupled online learning for neural network parameterizations in Earth System Models, aiming to reduce instabilities and biases by enabling the model to learn adaptively during simulations.

## Contribution

It proposes a novel coupled online learning approach that allows neural network parameterizations to adapt during simulations, improving stability and accuracy over traditional offline training methods.

## Key findings

- Coupled learning successfully recovers true parameterizations in Lorenz 96 model.
- The approach is adaptable to 3D cloud-resolving models.
- It addresses issues of instabilities and biases in online neural network parameterizations.

## Abstract

Over the last couple of years, machine learning parameterizations have emerged as a potential way to improve the representation of sub-grid processes in Earth System Models (ESMs). So far, all studies were based on the same three-step approach: first a training dataset was created from a high-resolution simulation, then a machine learning algorithm was fitted to this dataset, before the trained algorithm was implemented in the ESM. The resulting online simulations were frequently plagued by instabilities and biases. Here, coupled online learning is proposed as a way to combat these issues. Coupled learning can be seen as a second training stage in which the pretrained machine learning parameterization, specifically a neural network, is run in parallel with a high-resolution simulation. The high-resolution simulation is kept in sync with the neural network-driven ESM through constant nudging. This enables the neural network to learn from the tendencies that the high-resolution simulation would produce if it experienced the states the neural network creates. The concept is illustrated using the Lorenz 96 model, where coupled learning is able to recover the "true" parameterizations. Further, detailed algorithms for the implementation of coupled learning in 3D cloud-resolving models and the super parameterization framework are presented. Finally, outstanding challenges and issues not resolved by this approach are discussed.