The world most powerful particle accelerator, Large Hadron Collider (LHC) is design to collide particles at high energy in order to compare experimental observation from theoretical prediction made by the Standard Model of particle physics.
In order to analyse data output from collision events, it is necessary to identify particles through their signatures. (ie: energy deposition at different calorimeter, track information)
Examples of particles produced with high energy proton-proton collision are quarks or gluons. These particles hadronise and create a collimated spray of particles known as jets. They tend to deposit most of their energy in the hadronic calorimeter of the ATLAS detector (1 of 4 main collision points of LHC).
An attempt was done to build and train a Convolutional Neural Network (CNN) to identify jets created by W bosons. Simulated data provided by the UCL Physics and Astronomy Department is used to train and test performance of the model. The written code can be found in a jupyter notebook uploaded named "Jet_Tagging_CNN.ipynb"
A word document named "Jet_Tagging_Report.pdf" can be found with more details on the project.