SAIP2021

An Investigation of overtraining within Semi-Supervised Machine Learning Models in the search for heavy resonances at the LHC

30 Jul 2021, 12:45

15m

Potchefstroom Campus (North-West University)

Oral Presentation Track B - Nuclear, Particle and Radiation Physics Nuclear, Particle and Radiation Physics

Speaker

Benjamin Lieberman (University of Witwatersrand)

Description

When utilizing semi-supervised techniques in training machine learning models in the search for bosons at ATLAS, the overtraining of the model must be investigated. In particle physics internal fluctuations of the phase space and bias in training can cause semi-supervised models to label false signals within the phase space due to overfitting. The issue of false signal generation in semi-supervised models has not been fully analyzed and therefore utilizing a toy Monte Carlo model, the probability of such situations occurring can be quantified. This investigation of Zgamma resonances is performed using a pure background Monte Carlo sample. Through unique pure background samples extracted to mimic ATLAS data in a background-plus-signal region, multiple runs enable the probability of these fake signals occurring due to overtraining to be thoroughly investigated.

Apply to be considered for a student ; award (Yes / No)?

Yes

Level for award;(Hons, MSc, PhD, N/A)?

PhD

Presentation Materials

33 - Lieberman - An Investigation of overtraining within Semi-Supervised Machine Learning Models in the search for heavy resonances at the LHC.mp4

Peer reviewing

Paper

Accepted on 9 Dec 2021 by Aletta Prinsloo

Paper files:

• 33_Lieberman_saip2021_revised2.pdf
• 33_Lieberman_saip2021_comments2.pdf