Calorimetry with Deep Learning: Particle Classification, Energy Regression, and Simulation for High-Energy Physics

Gul Rukh Khattak

Abstract:

We present studies of the application of Deep Neural Networks and Convolutional Neural Networks for the classification, energy regression, and simulation of particles produced in high-energy particle collisions. We train cell-based Neural Nets that provide significant improvement in performance for particle classification and energy regression compared to feature-based Neural Nets and Boosted Decision
Trees, and Generative Adversarial Networks that provide reasonable modeling of several but not all shower features