Die DPG-Frühjahrstagung in Bonn musste abgesagt werden! Lesen Sie mehr ...

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

T: Fachverband Teilchenphysik

T 5: Machine Learning: QCD and electromagnetic showers

T 5.2: Vortrag

Montag, 30. März 2020, 16:45–17:00, H-HS III

Simulation of Extensive Air Showers with Deep Neural Networks — Steffen Hahn, •Marcel Köpke, and Markus Roth — Karlsruhe Institute of Technology, Institute for Nuclear Physics

The Pierre Auger Observatory uses CORSIKA to simulate extensive air showers. With growing incident particle energy it becomes computationally difficult to run the simulations due to increasing time complexity. Deep neural networks possess the ability to recognize patterns in an automatic way and are able to run on specialized, fast hardware like GPUs. Hence they are a good candidate to address run time issues while also offering the possibility to go beyond CORSIKA features like conditioning on meta parameters.