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

T: Fachverband Teilchenphysik

T 120: Data, AI, Computing 9 (generative models & simulation)

T 120.1: Vortrag

Freitag, 8. März 2024, 09:00–09:15, Geb. 30.34: LTI

Faster Simulations of Instrument Response Functions for Imaging Air Cherenkov Telescopes through Methods of Adaptive Sampling — •Tristan Gradetzke and Stefan Fröse — TU Dortmund University, Dortmund, Germany

Monte Carlo simulations of particle induced extensive air showers are of crucial importance to the analysis chain of data taken by Imaging Air Cherenkov Telescopes (IACTs). Besides for the training of particle classifiers and energy estimators, they are necessary to calculate the instrument response in the form of the Instrument Response Functions (IRFs). There usage however, comes at the extensive cost of computational resources. Therefore much effort has been made to this day, to make these simulations more efficient. This work, aims at investigating how to use them more efficiently for IRF calculations, thus reducing the amount needed. This is sought to be achieved by simulating only discrete points in energy and field of view, instead of continuous distributions currently used. The goal is to only sample the regions improving uncertainty and event statistics. The achieved uncertainties and event statistics are then compared to the standard approach. An outlook is given, on how methods of machine learning can be used to fasten the process even further.

Keywords: IACT; Intrument Response Function; Simulation; Adaptive Sampling