Heidelberg 2022 – wissenschaftliches Programm

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

T: Fachverband Teilchenphysik

T 53: Data Analysis, Information Technology and Artificial Intelligence 3

T 53.9: Vortrag

Dienstag, 22. März 2022, 18:15–18:30, T-H38

Using ML to analytically model the CMS detector response to jets — •Nils Gerber, Samuel Bein, and Peter Schleper — Universität Hamburg

Many applications in particle physics require an accurate modelling of the energy response of detectors to individual particles as well as jets. For example, unfolding, fast detector simulation, as well as certain background estimation techniques, all require some input related to the jet response. While typical models are constructed from established functional forms such as Crystal Ball functions fit to data distributions of the response, an alternative approach is explored where a DNN classifier is employed in order to arrive at a model which takes into account correlated dependencies in the response on the true jet energy, pseudorapidity, jet flavour, and other factors.