CosmoAI-AES is an inter-disciplinary research group at the Department of ICT and Natural Sciences at NTNU in Ålesund, working on gravitational lensing.
Professor Hans Georg Schaathun is a computer scientist working on simulation and machine learning
Dr Ben David Normann is a mathematical physicist working on cosmology
These pages present prototypes for two pieces of software.
CosmoSim is a simulator of gravitational lensing, supporting both an interactive and graphical user interface, and a command line interface for bulk generation of images.
droulette is not publicly available as of March 2026, but will be made public in the future. It provides the plumbing code to apply machine learning to images from CosmoSim.
Warning! These pages are under construction, and there will be errors and inconsistencies.
Using CosmoSim¶
Installation CosmoSim is a Python Package and can be installed with
pip install CosmoSim.CosmoSim GUI, which is run as
python -m CosmoSim.GUI.CosmoSim Demos exemplify intended workflows for testing the simulator, comparing simulation and lens models, and training and testing of machine learning models. Most of the demos use the API in Jupyter Notebook. Some depend on command line operation.
To understand the software, it will be necessary to have at least a superficial understanding of the Mathematical model.
Dokumentation¶
Other information¶
We presented the first prototype at ECMS in 2023 Schaathun et al., 2023, and a white paper on the overall approach later the same year, in Norwgian Schaathun et al., 2023. More recently we have made a survey on gravitational lensing Normann et al., 2025.
- Schaathun, H. G., Normann, B. D., Austnes, E. L., Ingebrigtsen, S., Remøy, S. W., & Runde, S. N. (2023). On the simulation of gravitational lensing in the roulette formalism. In E. Vicario, R. Bandinelli, V. Fani, & M. Mastroianni (Eds.), Proceedings of the 30th European Conference on Modelling and Simulation. ECMS - European Council for Modelling. 10.7148/2023-0141
- Schaathun, H. G., Normann, B. D., & Solev\aag-Hoti, K. (2023). Om \aa kartleggja mørk materie med maskinlæring. Norsk IKT-Konferanse for Forskning Og Utdanning, 2023(1). https://www.ntnu.no/ojs/index.php/nikt/article/view/5666
- Normann, B. D., Solev\aag-Hoti, K., & Schaathun, H. G. (2025). A short review on joint weak and strong cluster lens-mass reconstruction. The European Physical Journal Plus, 140(1175). 10.1140/epjp/s13360-025-07095-1