Authors:

Jack Patrick Rodgers

Date Created:

Not specified

Course Title:

Professor:

Not specified

About Paper:

In the Large Hadron Collider (LHC) at CERN, protons collide more than a million times per second. Pileup, which are interactions in the same or nearby proton bunch crossings in the accelerator, can be thought of as noise which affects many reconstructed physics variables such as the jet mass and jet transverse momentum. This noise results in worse resolution and as a consequence lower physics analysis sensitivity. Furthermore, pileup is expected to increase by a factor of 4 or 5 to be ~200 for the future LHC run that starts until 2029. Currently, an algorithm called PUPPI (Pileup Per Particle Identification) exists to mitigate pileup. A previous proof of concept study that uses a semi-supervised graph neural network for particle level pileup mitigation was tested using CMS fast simulation samples. The idea is to connect the training samples (labeled) and testing samples (unlabeled) as nodes in a graph using tracking and physics information. In addition, a dedicated masking technique was applied to reduce bias. The model was retrained using CMS full simulation which introduced more complexity in geometry and consequently graph neighbor construction. This increase in the complexity of geometry necessitated a more complex masking technique of particle labels. To use hyperparameter tuning in conjunction with these masking parameters, we introduce a bayesian optimization framework that aims to minimize a statistical performance metric of the physics performance observables for validation datasets. The current results from the Semi-Supervised Graph Neural Network outperform the baseline pileup mitigation algorithm PUPPI. Keywords: Deep Learning; Graph Neural Network; Physics

Source:

Purdue University / 2024

Topics:

No topics listed

Co-authors:

Jack Patrick Rodgers