Leveraging Machine Learning in the Search for New Bosons at the LHC and Other Resulting Applications - Master’s Dissertation

Aug 3, 2026

Upload by Miles

This dissertation examines semi-supervised machine learning for data generation in high-energy physics, with the goal of supporting searches for new bosons at the Large Hadron Collider. The core analysis builds a generative machine learning model to help identify resonances within the Zγ final-state background. Multiple Variational Auto-encoder (VAE) derivatives are developed, trained on a target Monte Carlo fast-simulated dataset, and evaluated via metrics and diagnostic plots to determine generation performance. The work also documents a related application of machine learning in COVID-19 crisis management.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages104

Reading Time4 h 22 min

AccessFree to download

Declaration of Authorship

Abstract

Acknowledgements

Research Output

Published Papers

Published or Soon to be Published Proceedings

UNIVERSITY OF THE WITWATERSRAND MASTER ’S DISSERTATION Leveraging Machine Learning in the Search for New Bosons at the LHC and Other Resulting Applications Author: Finn STEVENSON Supervisor: Prof. Bruce MELLADO A dissertation submitted in fulfillment of the requirements for the degree of Master of Science in the Wits Institute of Collider Particle Physics School of Physics September 11, 2023 iii Declaration of Authorship I, Finn STEVENSON, declare that this thesis titled,“Leveraging Machine Learning in the Search for New Bosons at the LHC and Other Resulting Applications” and the work presented in it are my own. I confirm that: • This work was done wholly or mainly while in candidature for a research degree at this University. • Where any part of this thesis has previously been submitted for a degree or any other qualification at this University or any other institution, this has been clearly stated. • Where I have consulted the published work of others, this is always clearly attributed. • Where I have quoted from the work of others, the source is always given. With the exception of such quotations, this thesis is entirely my own work. • I have acknowledged all main sources of help. • Where the thesis is based on work done by myself jointly with others, I have made clear exactly what was done by others and what I have contributed myself. Signed: 09/11/2023 Date: v “Our sole responsibility is to produce something smarter than we are; any problems beyond that are not ours to solve.” Ray Kurzweil vii UNIVERSITY OF THE WITWATERSRAND Abstract Science School of Physics Master of Science Leveraging Machine Learning in the Search for New Bosons at the LHC and Other Resulting Applications by Finn STEVENSON This dissertation focuses on the use of semi-supervised machine learning for data generation in high-energy physics, specifically to aid in the search for new bosons at the Large Hadron Collider. The overarching physics analysis for this work involves the development of a generative machine learning model to assist in the search for resonances in the Zγ final state background data. A number of Variational Auto-encoder (VAE) derivatives are developed and trained to be able to generate a chosen Monte Carlo fast simulated dataset. These VAE derivatives are then evaluated using chosen metrics and plots to assess their performance in data generation. Overall, this work aims to demonstrate the utility of semi-supervised machine learning techniques in the search for new resonances in high-energy physics. Additionally, a resulting application of the use of machine learning in COVID-19 crisis management was also documented. ix Acknowledgements I would like to sincerely thank my supervisor, Prof. Bruce Mellado, for his continued guidance, his sharing of knowledge and the opportunities he has afforded me throughout my master’s degree. Prof. Mellado has not only given me invaluable knowledge related to the fields of machine learning and particle physics but has taught me how to work under pressure in a way that elevates the quality of my work. I would also like to thank my fellow colleagues in the Wits physics department, especially Benjamin Lieberman, whose assistance and collaboration have been invaluable in completing this research. It is important to acknowledge and express gratitude to the various agencies from which I have been given funding to complete my research; the Africa-Canada Artificial Intelligence Data Modelling Consortium, the National Research Foundation and SA-CERN. I was able to travel to Geneva, Switzerland to work on my research at CERN as a result of the SA-CERN partnership. The experience of going to work at CERN as a member of the ATLAS experiment, one of the most advanced scientific collaborations in the world is something I will never forget. Lastly, thanks to my parents, Mark and Julie Stevenson for their continued support, sacrifices and encourageme

FAQs

What is the main purpose of the dissertation?

It focuses on using semi-supervised machine learning for data generation to support searches for new bosons at the LHC, including a generative modeling approach for resonance identification in Zγ background data.

What model type is used for the data generation task?

The dissertation develops and trains several Variational Auto-encoder (VAE) derivatives to generate a chosen Monte Carlo fast-simulated dataset.

How are the VAE derivatives evaluated?

They are assessed using chosen metrics and plots designed to quantify and visualize their performance in data generation.

Aug 3, 2026

Upload by Miles