AI for Collider Physics: Simulation, Inference, and the Search for New Physics

by Felix Ringer (YITP Stony Brook)

Tuesday 1 Dec 2026, 15:30 → 16:30 US/Eastern

Large Seminar Room

The growing complexity of modern collider experiments is transforming how physics analyses are performed, requiring new approaches to simulation, inference, and discovery in high-dimensional data. Recent advances in artificial intelligence and machine learning provide powerful tools to address these challenges by integrating event generation, statistical inference, and data analysis into unified computational frameworks. In this talk, I will review recent developments in AI-based physics analyses relevant to experiments at RHIC, the LHC, Jefferson Lab, and the future Electron-Ion Collider. I will discuss how generative models can improve collider simulations, how simulation-based inference enables the extraction of physical parameters directly from event-level data, and how unsupervised learning techniques can support searches for new physics signals. Together, these approaches point toward a shift from traditional simulation-driven workflows to data-driven inference pipelines that combine theoretical constraints with modern machine learning methods to improve precision measurements and expand discovery potential.