Prof. Dr. Christian Glaser, Principal Investigator at the Lamarr Institute for Machine Learning and Artificial Intelligence and professor at TU Dortmund University, has been awarded the 2026 Wallmark Prize by the Royal Swedish Academy of Sciences. He is being honored for his contributions to the radio detection of ultra-high-energy neutrinos and to the real-time analysis of measurement data using machine learning methods. The award underscores the growing role of artificial intelligence as a methodological component of data-intensive research. Glaser’s work at the Lamarr Institute exemplifies how AI helps open up new avenues of understanding for physical phenomena that were previously difficult to measure.
Research at the limits of what can be measured
The Wallmark Prize has been awarded since 1859 for scientific achievements that significantly advance science and technology. The focus is on work that opens up new experimental approaches to phenomena that were previously difficult to study. Glaser’s research addresses precisely this challenge. High-energy neutrinos are among the rarest measurable particles of all; their signatures are weak and appear only sporadically in large detectors. Glaser develops radio detection methods that make these signals accessible in large-scale experiments in the first place. His contributions range from physical modeling and the development of detection hardware to the analysis of complex measurement data—including in major international projects such as IceCube-Gen2 at the South Pole, a detector for measuring neutrinos in Antarctic ice, and RNO-G in Greenland, which uses radio waves to detect extremely high-energy particles. Glaser’s work is closely integrated into research in the field of astroparticle physics and exemplifies the Lamarr Institute’s activities in the Physics Area.
AI as a methodological driver of interdisciplinary research
Artificial intelligence plays an operational role within the experimental chain. Machine learning methods are used to identify rare particle signatures in highly noisy measurement data, classify events in real time, and reconstruct physical parameters from indirect measurements. This shifts the function of AI: it is not merely a tool for post-hoc analysis, but an integral part of data collection and interpretation. This is particularly relevant for detectors that continuously generate large data streams and in which relevant events account for only a fraction of the measurements. With the ERC-funded NuRadioOpt project, Glaser aims to systematically further develop these approaches and optimize detection strategies themselves using AI. For the Lamarr Institute, a central principle becomes evident here: Advances in AI are increasingly emerging through close interaction with complex application domains. Astroparticle physics exemplifies how algorithmic innovation and physical modeling interact to open up new approaches to phenomena that were previously difficult to measure.