Cuts Optimization and Machine Learning Models for Dark Photon Signal-Background Discrimination with the ATLAS Detector
Abstract/Contents
- Abstract
- Massless dark photons are theorized to mediate long-range forces between dark particles. They could lead to new and observable signatures in Higgs boson decays at the Large Hadron Collider (LHC). The discovery of dark photons would provide evidence of Beyond the Standard Model (BSM) physics and could help solve small-scale structure formation problems in cosmology. In this thesis, we consider the ZH production mechanism to search for a newly-predicted decay of the Higgs boson into a photon (γ) and a massless dark photon (γD) with a target final state of Z(→l+l−)H(→γγD). We use Monte-Carlo simulation samples corresponding to the total integrated luminosity of pp collisions at √s=13 TeV collected by the ATLAS detector in 2015-2018. First, we implement rectangular cuts optimization to obtain a baseline signal region (SR) selection. Next, we explore the use of boosted decision trees (BDTs) and neural networks to train a classifier for signal-background discrimination. BDTs using the Gradient Boosting algorithm with the xgBoost implementation yield the best performance with 96% accuracy and an area under the Receiver Operating Characteristic (ROC) curve (AUC) of 0.97. We then compare the performance of the BDT to the rectangular cuts optimization through evaluating the approximate median significance (AMS) metric on the simulation samples. We find that cutting on the BDT output score distribution has a much higher AMS than the rectangular cuts optimization (∼10 compared to ∼1), thus significantly improving the sensitivity of this analysis for dark matter detection.
Description
| Type of resource | text |
|---|---|
| Date created | May 16, 2021 |
| Date modified | April 7, 2023 |
| Publication date | June 1, 2021 |
Creators/Contributors
| Author | Hofgard, Elyssa |
|---|---|
| Degree granting institution | Stanford University, Department of Physics |
| Thesis advisor | Tompkins, Lauren |
| Thesis advisor | Schwartzman, Ariel |
Subjects
| Subject | particle physics |
|---|---|
| Subject | dark photons |
| Subject | machine learning |
| Subject | ATLAS experiment |
| Subject | data analysis |
| Genre | Text |
| Genre | Thesis |
Bibliographic information
| Related item | |
|---|---|
| Location | https://purl.stanford.edu/pb798fj0435 |
Access conditions
- Use and reproduction
- User agrees that, where applicable, content will not be used to identify or to otherwise infringe the privacy or confidentiality rights of individuals. Content distributed via the Stanford Digital Repository may be subject to additional license and use restrictions applied by the depositor.
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 4.0 International license (CC BY-NC).
Preferred citation
- Preferred citation
- Hofgard, Elyssa. (2021). Cuts Optimization and Machine Learning Models for Dark Photon Signal-Background Discrimination with the ATLAS Detector. Stanford Digital Repository. Available at: https://purl.stanford.edu/pb798fj0435
Collection
Undergraduate Theses, Department of Physics
View other items in this collection in SearchWorksContact information
- Contact
- ehofgard@stanford.edu
Also listed in
Loading usage metrics...