Detail of the student project

Topic:Hledání leptoquarků pomocí strojového učení v datech z CERN ATLAS experimentu
Department:Katedra kybernetiky
Supervisor:doc. André Sopczak
Announce as:Diplomová práce, Bakalářská práce, Semestrální projekt
Description:At the Large Hadron Collider (LHC) at CERN protons are collided and the collisions are recorded by the ATLAS detector.
A motivation to construct the LHC has been the search for new particles. Such new particles could be Leptoquarks. Leptoquarks
are predicted by several theories, but so far have not been detected in the recorded data. Their mass is unknown.
Leptoquark signal events have been simulated, together with other events resulting from background reactions.
The task is to recognize these simulated signal events automatically from other (background) events using the techniques of machine learning and possibly deep learning.

1. Get familiar with the data and basic principles of searching for elementary particles in high-energy physics.
2. Get familiar with the existing implementation of classifiers to separate events of interest from the background, using high-level features and classical machine-learning techniques.
3. Design and implement a classifier based on high-level features using either classical machine learning or deep-learning or both. Evaluate its performance on simulated data and compare with existing approaches for a large Leptoquark mass range.

4. Apply the classifier on recorded real data and compare the number of detected events with the theoretical predictions.

This project will be supervised by doc. Andre Sopczak, UTEF ČVUT.
Bibliography:[1] (“learn more”) An introduction to the ATLAS experiment for the public
[2] Dan Guest et al. Deep Learning and its application to LHC Physics. Annu. Rev. Nucl. Part. Sci. 2018, 68:1-22
[3] Pierre Baldil et al: Searching for Exotic Particles in High-Energy Physics. arXiv:1402.4735
[4] ATLAS Collaboration: Search for pair production of third-generation scalar leptoquarks decaying into a top quark and a tau-lepton in pp collisions at √s = 13 TeV with the ATLAS detector. JHEP 06 (2021) 179
[5] A.Sopczak, Searches for Leptoquarks with the ATLAS Detector, arXiv:2107.10094
[6] R.Duda, P.Hart, D.Stork: Pattern classification. Willey-Interscience, 2000
[7] Goodfellow, Bengio, Courville: Deep learning. MIT Press. 2016
Responsible person: Petr Pošík