Speaker
Benjamin Lieberman
(University of Witwatersrand)
Description
In resonance searches for new physics, machine learning techniques are used to classify signal from background events. When using machine learning classifiers it is necessary to measure the amount of background events being incorrectly labelled as signal events. In this research the Zγ→(ℓ+ℓ−)γ final state dataset focusing around 150GeV centre of mass is used. A Wasserstein Generative Adversarial Network is used as a generative model and a semi-supervised DNN is used as a classifier. This study provides a methodology and the results of the measurement of false signals generated during the training of semi-supervised DNN classifiers.
Level for award;(Hons, MSc, PhD, N/A)?
PhD
| Apply to be considered for a student ; award (Yes / No)? | Yes |
|---|
Primary author
Benjamin Lieberman
(University of Witwatersrand)
Co-authors
Bruce Mellado
(University of the Witwatersrand)
Xifeng Ruan
(University of the Witwatersrand)
Finn Stevenson
(University of the Witwatersrand)
