High-energy physics (HEP) strives for understanding nature’s fundamental building blocks and the interactions between them. Large amounts of high-dimensional data, collected for example with the experiments at the Large Hadron Collider (LHC) at CERN in Switzerland are compared to theoretical predictions based on quantum field theory. Modern Machine Learning (ML) brought a lot of new ideas and improvements to HEP in recent years. On the experimental side, this includes data collection, particle reconstruction and selection as well as subsequent analysis. On the theoretical side, ML contributed a lot to better and faster simulation and parameter inference.

Overall, ML has not only led to significant improvements of existing algorithms, but also to new ideas for previously considered impossible-to-solve problems. Given the broad range of applicability of ML in HEP, almost all different types of objectives (like regression, classification, generation) on various different types of data (like tabular, point clouds, graphs) are being explored. In this SIG, we bring together experts from various domains of particle physics to use ML to improve our understanding of Nature.

Members

• Robert Schöfbeck (HEPHY)
• Wolfgang Waltenberger (HEPHY)
• Andreas Ipp (TU Vienna)
• Postdocs in CMS and HEP-ML group at HEPHY (currently 5 people)

Activities

Physics, and especially High-Energy Physics is a key part of the successful proposal for an AI:Factory in Austria. I was involved in drafting that section of the proposal. https://eurohpc-ju.europa.eu/eurohpc-ju-selects-additional-ai-factories-strengthen-europes-ai-leadership-2025-03-12_en. Physics is one of the main pillars for research at the AI:AT factory.

Members of the SIG successfully participated in the FAIR Universe Higgs Uncertainty Challenge. As of April 7th, we won the public phase of the challenge (we were leading the leader board on March 14, when that phase concluded) and we are now awaiting the results of the final, high-statistics evaluation run. We were already told in private that we are in the top 3, so we will receive some prize money. I expect us to win and we should be notified within the next days.

Robert Schöfbeck’s FWF Proposal on using machine-learning techniques for LHC data analysis was approved: https://www.fwf.ac.at/forschungsradar/10.55776/PAT7453824

Andreas Ipp’s FWF Proposal on using machine-learning techniques for improving Lattice QCD simulations was approved: https://www.fwf.ac.at/forschungsradar/10.55776/PAT3667424