Electron Momentum Reconstruction using Supervised Machine Learning at the LHCb

Bébr, Mikuláš (2022) Electron Momentum Reconstruction using Supervised Machine Learning at the LHCb. Bachelor's Thesis, Physics.

Preview	Text bPHYS_2022_BebrM.pdf Download (2MB) | Preview
	Text toestemming.pdf Restricted to Registered users only Download (125kB)

Abstract

This research focuses on improving the momentum reconstruction method employed at the LHCb by using supervised machine learning. An overview of the principle of lepton universality violation and its implication is provided. A performance comparison between the Decision Tree regression algorithm and Polynomial regression using the least-squares method is given as well as the drawbacks of each method. The main result is the plot of the residual sum of the reconstructed momentum and the true value of the momentum from simulated data of the LHCb. The research concludes that although the supervised learning machine algorithms do provide a general decrease in the standard deviation of the residual sum, it does inflate the distribution around 0, effectively reducing the sharpness of the peak of the distribution.

Item Type:	Thesis (Bachelor's Thesis)
Supervisor name:	Veghel, M.C. van and Onderwater, C.J.G.
Degree programme:	Physics
Thesis type:	Bachelor's Thesis
Language:	English
Date Deposited:	13 Jul 2022 11:09
Last Modified:	13 Jul 2022 11:09
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/27712

Actions (login required)

View Item