Particles and Fields Seminar — HEP seminar
Graph Neural Networks for Particle Physics
Mrs. Nathalie Soybelman
WIS
Abstract
In LHC experiments, particle collision data is recorded by complex detectors storing up to 6 GB of data per second.
Given this data’s complexity, cutting-edge machine learning is not just a tool but the essential pathway for its interpretation, simulation, and a comprehensive understanding of its implications regarding the fundamental laws of physics.
Graphs offer a natural representation of particle physics data. Consequently, we employ graph neural networks to address this multitude of challenges.
In this talk, I will give an overview of the current efforts in employing graph neural networks for event simulation, particle reconstruction, and physics analysis.
Given this data’s complexity, cutting-edge machine learning is not just a tool but the essential pathway for its interpretation, simulation, and a comprehensive understanding of its implications regarding the fundamental laws of physics.
Graphs offer a natural representation of particle physics data. Consequently, we employ graph neural networks to address this multitude of challenges.
In this talk, I will give an overview of the current efforts in employing graph neural networks for event simulation, particle reconstruction, and physics analysis.