BEGIN:VCALENDAR VERSION:2.0 PRODID:-//linuxsoftware.nz//NONSGML Joyous v1.4//EN BEGIN:VEVENT SUMMARY:Optimization of a Transformer-Based Model for Flavour-Tagging in t he ATLAS Experiment DTSTART:20251125T160000Z DTEND:20251125T180000Z DTSTAMP:20260906T150906Z UID:1628c2d7-f57e-4557-8509-fad8c2cdb844 SEQUENCE:4 CREATED:20251121T092456Z DESCRIPTION:The Standard Model is the most widely accepted theory in the f ield of particle physics. In order to test and find some of the missing pi eces of the theory\, several experiments are being made around the world. One of the main drivers of experimental particle physics research are coll ider experiments\, where particles are accelerated and carefully made to c ollide at specific regions with detectors ready to collect the products of the interaction. This data is later treated and the events reconstructed so that it is possible to learn about the phenomena occurring near the col lision point. This complex task has several stages\, one of which is calle d jet flavour-tagging\, that aims to identify the flavour of the particles that originated jets after the collision. Machine learning models are emp loyed with this task due to its complexity derived from the large amounts of data collected from each collision. One such model\, created by the ATL AS Collaboration\, called GN2\, is studied in this work with a focus on th e input variables it receives. In this thesis\, different versions of the model were created with different input variables to test their importance to the model’s performance. Additionally\, a study was conducted to dir ectly test the impact of these input variables on the model’s decision-m aking\, and the results suggest that a group of these variables related to the number of particle interactions with the detector may have a limited effect on the model. LAST-MODIFIED:20251124T114800Z LOCATION:Online URL:http://df.vps.tecnico.ulisboa.pt/en/events/optimization-of-a-transform er-based-model-for-flavour-tagging-in-the-atlas-experiment/ X-ALT-DESC;FMTTYPE=text/html:

The Standard Model is the most widely accepted theory in the field of particle physics. In or der to test and find some of the missing pieces of the theory\, several ex periments are being made around the world. One of the main drivers of expe rimental particle physics research are collider experiments\, where partic les are accelerated and carefully made to collide at specific regions with detectors ready to collect the products of the interaction.

Thi s data is later treated and the events reconstructed so that it is possibl e to learn about the phenomena occurring near the collision point. This co mplex task has several stages\, one of which is called jet flavour-tagging \, that aims to identify the flavour of the particles that originated jets after the collision. Machine learning models are employed with this task due to its complexity derived from the large amounts of data collected fro m each collision.


One such model\, created by the ATLAS Col laboration\, called GN2\, is studied in this work with a focus on the inpu t variables it receives. In this thesis\, different versions of the model were created with different input variables to test their importance to th e model’s performance. Additionally\, a study was conducted to directly test the impact of these input variables on the model’s decision-making\ , and the results suggest that a group of these variables related to the n umber of particle interactions with the detector may have a limited effect on the model.

END:VEVENT END:VCALENDAR