info@atmorep.org
AtmoRep
Large Scale Representation Learning of Atmospheric Data
The AtmoRep project asks if one can train one neural network that represents and describes all atmospheric dynamics. AtmoRep's ambition is hence to demonstrate that the concept of large-scale representation learning, whose principle feasibility and potential was established by large language models such as GPT-3, is also applicable to scientific data and in particular to atmospheric dynamics. The project is enabled by the large amounts of atmospheric observations that have been made in the past as well as advances on neural network architectures and self-supervised learning that allow for effective training on petabytes of data. Eventually, we aim to train on all of the ERA5 reanalysis and, furthermore, fine tune on observational data such as satellite measurements to move beyond the limits of reanalyses.
News
- March 2023: Christian presented AtmoRep at the IntelliAQ workshop in Cologne. We had some lively discussions on where and how machine learning will be best used in atmospheric science in the future.
- February 2023: Christian gave a talk at the NSF AI Institute for Research on Trustworthy AI in Weather, Climate, and Coastal Oceanography
- Atmorep has become part of the newly established WestAI service centre which aims at enhancing and accelerating large-scale deep learning models for language, image and weather data processing.
- December 2022: Christian presented an online poster at AGU on Tuesday, Dec. 13th
Publications
Journal publications
- S. Hoffmann and C. Lessig. Atmodist: Self-supervised representation learning for atmospheric dynamics. Environmental Data Science, 2:e6, 2023.
Talks
- C. Lessig, IntelliAQ workshop, March 2023; Slides.
- C. Lessig, NSF AI Institute for Research on Trustworthy AI in Weather, Climate, and Coastal Oceanography, February 2023; Slides.
- C. Lessig, Google Research, October 2022; Slides.
- I. Luise, DWD KI Forum, September 2022.
Poster
- C. Lessig et al., American Geophysical Union Annual Meeting 2022; Poster.
- C. Lessig et al., ECMWF Workshop on Machine Learning, Nov. 2022; Poster.
- I. Luise et al., Italian Society for Climate Sciences Annual Conference 2022.