# Climate-Invariant Machine Learning

arXiv: https://arxiv.org/abs/2112.08440 · HTML: https://mlsanity.com/paper/2112.08440/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 128 references that are in the map.
- First posted on arXiv 14 Dec 2021.

## Builds on (most cited first, 17 of 17)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems](https://mlsanity.com/paper/1603.04467/) (2016)
- [Cyclical Learning Rates for Training Neural Networks](https://mlsanity.com/paper/1506.01186/) (2015)
- [A Survey of Unsupervised Deep Domain Adaptation](https://mlsanity.com/paper/1812.02849/) (2018)
- [The (Un)reliability of saliency methods](https://mlsanity.com/paper/1711.00867/) (2017)
- [Deep learning to represent sub-grid processes in climate models](https://mlsanity.com/paper/1806.04731/) (2018)
- [A review of domain adaptation without target labels](https://mlsanity.com/paper/1901.05335/) (2019)
- [Recent Advances in Domain Adaptation for the Classification of Remote Sensing Data](https://mlsanity.com/paper/2104.07778/) (2021)
- [Machine Learning for the Geosciences: Challenges and Opportunities](https://mlsanity.com/paper/1711.04708/) (2017)
- [Will Artificial Intelligence supersede Earth System and Climate Models?](https://mlsanity.com/paper/2101.09126/) (2021)
- [A Fortran-Keras Deep Learning Bridge for Scientific Computing](https://mlsanity.com/paper/2004.10652/) (2020)
- [Towards Physically-consistent, Data-driven Models of Convection](https://mlsanity.com/paper/2002.08525/) (2020)
- [Cumulo: A Dataset for Learning Cloud Classes](https://mlsanity.com/paper/1911.04227/) (2019)
- [Combining crowd-sourcing and deep learning to explore the meso-scale organization of shallow convection](https://mlsanity.com/paper/1906.01906/) (2019)

## Built on by (most cited first, 24 of 24)

- [ACE: A fast, skillful learned global atmospheric model for climate prediction](https://mlsanity.com/paper/2310.02074/) (2023)
- [ACE2: Accurately learning subseasonal to decadal atmospheric variability and forced responses](https://mlsanity.com/paper/2411.11268/) (2024)
- [Can AI weather models predict out-of-distribution gray swan tropical cyclones?](https://mlsanity.com/paper/2410.14932/) (2024)
- [Drift-Resilient TabPFN: In-Context Learning Temporal Distribution Shifts on Tabular Data](https://mlsanity.com/paper/2411.10634/) (2024)
- [Explaining the physics of transfer learning a data-driven subgrid-scale closure to a different turbulent flow](https://mlsanity.com/paper/2206.03198/) (2022)
- [Multiscale Neural Operator: Learning Fast and Grid-independent PDE Solvers](https://mlsanity.com/paper/2207.11417/) (2022)
- [Robustness of AI-based weather forecasts in a changing climate](https://mlsanity.com/paper/2409.18529/) (2024)
- [Deep Learning Based Cloud Cover Parameterization for ICON](https://mlsanity.com/paper/2112.11317/) (2021)
- [CERA: A Framework for Improved Generalization of Machine Learning Models to Changed Climates](https://mlsanity.com/paper/2509.00010/) (2025)
- [Navigating the Noise: Bringing Clarity to ML Parameterization Design with O(100) Ensembles](https://mlsanity.com/paper/2309.16177/) (2023)
- [Extreme Event Prediction with Multi-agent Reinforcement Learning-based Parametrization of Atmospheric and Oceanic Turbulence](https://mlsanity.com/paper/2312.00907/) (2023)
- [Lightning-Fast Convective Outlooks: Predicting Severe Convective Environments with Global AI-based Weather Models](https://mlsanity.com/paper/2406.09474/) (2024)
- [Multi-scale decomposition of sea surface height snapshots using machine learning](https://mlsanity.com/paper/2409.17354/) (2024)
- [Forecasting the Future with Yesterday's Climate: Temperature Bias in AI Weather and Climate Models](https://mlsanity.com/paper/2509.22359/) (2025)
- [The Equilibrium Response of Atmospheric Machine-Learning Models to Uniform Sea Surface Temperature Warming](https://mlsanity.com/paper/2510.02415/) (2025)
- [Towards Physically Consistent Deep Learning For Climate Model Parameterizations](https://mlsanity.com/paper/2406.03920/) (2024)
- [Enforcing Equity in Neural Climate Emulators](https://mlsanity.com/paper/2406.19636/) (2024)
- [Simple and Robust Forecasting of Spatiotemporally Correlated Small Earth Data with A Tabular Foundation Model](https://mlsanity.com/paper/2510.08920/) (2025)
- [Machine Learning for Stochastic Parametrisation](https://mlsanity.com/paper/2402.09471/) (2024)
- [Conditional updates of neural network weights for increased out of training performance](https://mlsanity.com/paper/2512.03653/) (2025)
- [Accelerating Optimization and Machine Learning through Decentralization](https://mlsanity.com/paper/2604.19518/) (2026)
- [No Epoch Like the Present: Robust Climate Emulation Requires Out-of-Distribution Generalisation](https://mlsanity.com/paper/2605.22248/) (2026)
- [Position: AI Must Become Planet-Centered, Not Just Human-Centered](https://mlsanity.com/paper/2606.13704/) (2026)
- [Learning Prognostic Variables for AI Convective Parameterizations via Symbolic Distillation](https://mlsanity.com/paper/2609.24882/) (2026)

## Abstract (arXiv)

Projecting climate change is a generalization problem: we extrapolate the recent past using physical models across past, present, and future climates. Current climate models require representations of processes that occur at scales smaller than model grid size, which have been the main source of model projection uncertainty. Recent machine learning (ML) algorithms hold promise to improve such process representations, but tend to extrapolate poorly to climate regimes they were not trained on. To get the best of the physical and statistical worlds, we propose a new framework - termed "climate-invariant" ML - incorporating knowledge of climate processes into ML algorithms, and show that it can maintain high offline accuracy across a wide range of climate conditions and configurations in three distinct atmospheric models. Our results suggest that explicitly incorporating physical knowledge into data-driven models of Earth system processes can improve their consistency, data efficiency, and generalizability across climate regimes.