Quantum Generative Adversarial Networks for High Energy Physics Simulations

Rey Guadarrama, Sergei Gleyzer, Konstantin Matchev, Katia Matcheva, Kyoungchul Kong, Gopal Ramesh Dahale, Mariia Baidachna, Haydee Hernández-Arellano, Isabel Pedraza

Keywords: Quantum Computing, High Energy Physics, Quantum Machine Learning, Quantum Generative Adversarial Networks

TL;DR: This paper presents the implementation of a Quantum Generative Adversarial Network to generate gluon-initiated jet images from ECAL detector data, showcasing high fidelity in replicating energy deposits.

Abstract:

The potential for quantum computing to offer significant advantages over classical computing makes it a promising approach for exploring alternative future methods in High Energy Physics (HEP) simulations. This work presents the implementation of a Quantum Generative Adversarial Network (qGAN) to generate gluon-initiated jet images from ECAL detector data, a task crucial for high-energy physics simula-
tions at the Large Hadron Collider (LHC). The results demonstrate high fidelity in replicating energy deposit patterns and preserving the implicit training data features. This study marks the first step toward generating multi-channel pictures
and quark-initiated jet images.

Submission Number: 55