Particle Cloud Generation with Message Passing Generative Adversarial Networks

Autor(es):Mostrar menos -

Kansal, Raghav ; Duarte, Javier ; Su, Hao ; Orzari, Breno ; Tomei, Thiago ; Pierini, Maurizio ; Touranakou, Mary ; Vlimant, Jean-Roch ; Ranzato, M ; Beygelzimer, A ; Dauphin, Y ; Liang, PS ; Vaughan, JW

• Tipo de documento: Artigo Científico
• Fonte: ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 30 (NIPS 2017); v. 34, p. 14-pg., 2021-01-01.

Resumo

In high energy physics (HEP), jets are collections of correlated particles produced ubiquitously in particle collisions such as those at the CERN Large Hadron Collider (LHC). Machine learning (ML)-based generative models, such as generative adversarial networks (GANs), have the potential to significantly accelerate LHC jet simulations. However, despite jets having a natural representation as a set of particles in momentum-space, a.k.a. a particle cloud, there exist no generative models applied to such a dataset. In this work, we introduce a new particle cloud dataset (JetNet), and apply to it existing point cloud GANs. Results are evaluated using (1) 1-Wasserstein distances between high- and low-level feature distributions, (2) a newly developed Frechet ParticleNet Distance, and (3) the coverage and (4) minimum matching distance metrics. Existing GANs are found to be inadequate for physics applications, hence we develop a new message passing GAN (MPGAN), which outperforms existing point cloud GANs on virtually every metric and shows promise for use in HEP. We propose JetNet as a novel point-cloud-style dataset for the ML community to experiment with, and set MPGAN as a benchmark to improve upon for future generative models. Additionally, to facilitate research and improve accessibility and reproducibility in this area, we release the open-source JETNET Python package with interfaces for particle cloud datasets, implementations for evaluation and loss metrics, and more tools for ML in HEP development. (AU)

• Processo FAPESP: 19/16401-0 - Aprendizado de máquina para simulação de colisões de física de altas energias com o detector CMS
• Beneficiário: Breno Orzari
• Modalidade de apoio: Bolsas no Exterior - Estágio de Pesquisa - Mestrado
• Processo FAPESP: 18/25225-9 - Centro de Pesquisa e Análise de São Paulo
• Beneficiário: Sergio Ferraz Novaes
• Modalidade de apoio: Auxílio à Pesquisa - Projetos Especiais
• Processo FAPESP: 18/01398-1 - Busca por nova física no experimento CMS do Large Hadron Collider
• Beneficiário: Breno Orzari
• Modalidade de apoio: Bolsas no Brasil - Mestrado