Anomaly detection techniques are playing an increasingly central role in the search for new physics at the Large Hadron Collider (LHC), where the rarity and unpredictability of signals beyond the Standard Model challenge traditional analysis strategies. In this talk, I will present recent developments in machine learning–based anomaly detection for high-energy physics, with a focus on model-agnostic approaches that aim to uncover unexpected signatures in collision data. I will discuss how these algorithms are currently being used to probe the Standard Model at the LHC, and how next-generation and future searches might be deploying such algorithms, from FPGA-accelerated algorithms to quantum computing strategies.
Biosketch:
Dr. Benedikt Maier is an Eric and Wendy Schmidt AI in Science Postdoctoral Fellow at Imperial College London, leading the EPIGRAPHY network to develop deep learning for real-time edge computing in LHC experiments. As co-convenor of the CMS Exotics group, he advances searches for new physics and dark matter while innovating in machine learning for data analysis and large-scale computing. His leadership in managing CMS’s global data grid and his research achievements earned him the CMS Young Researcher Prize.