Identifying dark jets with Dynamic Graph Convolutional Neural Network

Adviser: Yevgeny Kats

Student: נועם טל

Abstract

In this report I will present an implementation as supervised deep neural network to identify dark jets. The network that I will use is the ParticleNet, a customized neural network architecture using Dynamic Graph Convolutional Neural Network which is a convolutional neural network that operates on particle clouds, similar to the concept of point clouds in computer vision, rather than analyze jet images. This approach considers a jet as an unordered set of particles, a “particle cloud”, that respects the permutation symmetry and has been proven to be efficient when dealing with particle jets tagging.
Advertised on Feb. 18, 2021
Abstract added on Feb. 18, 2021
Abstract approved on Feb. 18, 2021 by Kovetz, Ely (kovetz)