SMuK 2023 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

T: Fachverband Teilchenphysik

T 86: ML Methods IV

T 86.1: Vortrag

Mittwoch, 22. März 2023, 17:30–17:45, HSZ/0405

EPiC-GAN: Equivariant Point Cloud Generation for Particle Jets — •Erik Buhmann — Institut für Experimentalphysik, Universität Hamburg

With current and future high-energy collider experiments' vast data-collecting capabilities comes an increasing demand for computationally efficient simulations. Generative machine learning models allow fast event generation, yet so far are largely constrained to fixed data and detector geometries. We introduce the Deep Sets-based equivariant point cloud generative adversarial network (EPiC-GAN) for the generation of point clouds with variable cardinality -- a flexible data structure optimal for collider events such as jets. The generator and discriminator utilize multiple EPiC layers with an interpretable global latent vector and do not rely on pairwise information sharing between particles, leading to a significant speed-up over graph- and transformer-based approaches. We show that our GAN scales well to large particle multiplicities and achieves high generation fidelity for gluon, light quark, and top jets.