AI research for more precise particle-physics simulations receives funding
zukunft.niedersachsen funds Göttingen project led by Dr Timo Janßen
The University of Göttingen has been awarded around 1.8 million euros for a new AI research group in the programme “KI-Forschungsgruppen Niedersachsen” funded by zukunft.niedersachsen. The five-year project “Production-ready AI for unbiased simulation in high energy physics” will be led by Dr Timo Janßen at the Institute for Theoretical Physics. It aims to make computer simulations of high-energy particle collisions much faster while preserving the scientific accuracy that is essential for research at CERN’s Large Hadron Collider and its upcoming high-luminosity phase.Particle physicists use such simulations to compare theoretical predictions with experimental data and to search for new phenomena. However, producing the enormous number of reliable simulated events needed for modern experiments requires substantial computing power. The Göttingen team will develop trustworthy AI methods that speed up key steps in these simulations without distorting the results. A central deliverable will be a modular software library that makes the new methods available through stable interfaces, documentation and benchmarks, so that established simulation tools can use them in practice.
“The next generation of collider experiments will confront us with unprecedented amounts of data,” says Janßen. “Our goal is to use AI not as a black box, but as a reliable accelerator for physics simulations: faster calculations, lower computing costs and, most importantly, results that researchers can trust.” The project is closely linked to the Campus-Institute Data Science (CIDAS); Janßen previously received a CIDAS Fellowship, which helped him secure this new funding. The group will also draw on computing expertise and infrastructure from the GWDG, including resources connected to NHR-Nord.
“This success emphasizes the strength of physics research in Göttingen and our close connection to data science and high-performance computing. Dr. Janssen’s project demonstrates how methodological innovation opens up new possibilities for answering fundamental questions about nature,” emphasizes Prof. Dr. Sarah Köster, dean of the Faculty of Physics. “We are extremely pleased to welcome Dr Janßen and his junior research group to the Institute for Theoretical Physics,” says Prof. Steffen Schumann, Director of the Institute for Theoretical Physics. “By developing innovative AI-based simulation methods for accelerator experiments, Dr Janßen is an excellent fit for strengthening our research in theoretical particle physics and at the same time opens up new opportunities for collaboration on the Göttingen Campus.”