Speaker
Description
The DeepTau tau identification algorithm, based on Deep Neural Network techniques, has been developed to reduce the fraction of jets, muons and electrons misidentified as hadronically decaying tau leptons by the Hadron-plus-strip algorithm. Its recently deployed version 2.5 for Run3 has brought several improvements to the existing algorithm, e.g. the addition of domain adaptation to reduce data-MC discrepancies in the high-confidence region of the tagger. The resulting model delivers a reduced mis-identification rate at a given efficiency by 10-50% across the regions of interest and, thus, sets a new improved baseline for the tau identification task. The talk will focus on DeepTau v2.5, the comparison with the previous version (v2.1) and its calibration using pp collisions. Corrections to improve data modeling are also shown.
| Alternate track | 14. Computing, AI and Data Handling |
|---|---|
| I read the instructions above | Yes |
