Erlangen 2026 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

T: Fachverband Teilchenphysik

T 93: Data, AI, Computing, Electronics VIII

T 93.1: Talk

Friday, March 20, 2026, 09:00–09:15, KH 00.024

Electromagnetic Shower Shape Correction using Normalizing Flows — •Marius Melcher, Arno Straessner, and Asma Hadef — Technische Universität Dresden

The interplay between data and simulation is essential for precision measurements at the ATLAS experiment. The Geant4-based simulation of electromagnetic showers in the calorimeter, however, exhibits notable discrepancies with data, particularly in derived shower-shape variables used for electron and photon identification. These mismatches can impact the efficiency and background rejection of identification algorithms, making corrections to simulated showers crucial for physics analyses.

This talk presents a machine-learning-based approach using normalizing flows to correct shower-shape variables in simulation. This model is applied to electron shower-shapes measured by ATLAS during LHC Run 3. Resulting improvements in data-simulation agreement are discussed.

Keywords: electrons; normalizing flows; shower shape; ATLAS