Enregistré dans:
Détails bibliographiques
Auteurs principaux: Delgado, Andrea, Venegas-Vargas, Diego, Huynh, Adam, Carroll, Kevon
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2410.12650
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866913548937986048
author Delgado, Andrea
Venegas-Vargas, Diego
Huynh, Adam
Carroll, Kevon
author_facet Delgado, Andrea
Venegas-Vargas, Diego
Huynh, Adam
Carroll, Kevon
contents Generative modeling for high-resolution images in Liquid Argon Time Projection Chambers (LArTPC), used in neutrino physics experiments, presents significant challenges due to the complexity and sparsity of the data. This work explores the application of quantum-enhanced generative networks to address these challenges, focusing on the scaling of models to handle larger image sizes and avoid the often encountered problem of mode collapse. To counteract mode collapse, regularization methods were introduced and proved to be successful on small-scale images, demonstrating improvements in stabilizing the training process. Although mode collapse persisted in higher-resolution settings, the introduction of these techniques significantly enhanced the model's performance in lower-dimensional cases, providing a strong foundation for further exploration. These findings highlight the potential for quantum-enhanced generative models in LArTPC data generation and offer valuable insights for the future development of scalable hybrid quantum-classical solutions in nuclear and high-energy physics.
format Preprint
id arxiv_https___arxiv_org_abs_2410_12650
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Towards Designing Scalable Quantum-Enhanced Generative Networks for Neutrino Physics Experiments with Liquid Argon Time Projection Chambers
Delgado, Andrea
Venegas-Vargas, Diego
Huynh, Adam
Carroll, Kevon
Quantum Physics
High Energy Physics - Experiment
Generative modeling for high-resolution images in Liquid Argon Time Projection Chambers (LArTPC), used in neutrino physics experiments, presents significant challenges due to the complexity and sparsity of the data. This work explores the application of quantum-enhanced generative networks to address these challenges, focusing on the scaling of models to handle larger image sizes and avoid the often encountered problem of mode collapse. To counteract mode collapse, regularization methods were introduced and proved to be successful on small-scale images, demonstrating improvements in stabilizing the training process. Although mode collapse persisted in higher-resolution settings, the introduction of these techniques significantly enhanced the model's performance in lower-dimensional cases, providing a strong foundation for further exploration. These findings highlight the potential for quantum-enhanced generative models in LArTPC data generation and offer valuable insights for the future development of scalable hybrid quantum-classical solutions in nuclear and high-energy physics.
title Towards Designing Scalable Quantum-Enhanced Generative Networks for Neutrino Physics Experiments with Liquid Argon Time Projection Chambers
topic Quantum Physics
High Energy Physics - Experiment
url https://arxiv.org/abs/2410.12650