Speaker
Description
Machine learning (ML) has become an essential tool in high-energy physics (HEP). It has changed the way how data is processed across every stage of a collider experiment. Today, ML techniques power real-time trigger decisions, detector simulation, object reconstruction, and final statistical inference. This talk provides an overview of ML applications in HEP. I will focus mainly on collider physics, and will highlight the shift toward end-to-end deep learning. By training directly on raw detector readouts, these ML architectures bypass traditional feature engineering, minimize bias, and unlock sensitivity in challenging signatures where standard algorithms fall short. We discuss how these modern ML techniques continuously expand the physics reach and discovery
potential of collider data.