Particle Graph Autoencoders and Differentiable, Learned Energy Mover's Distance

Abstract

Autoencoders have useful applications in high energy physics in anomaly detection, particularly for jets - collimated showers of particles produced in collisions such as those at the CERN Large Hadron Collider. We explore the use of graph-based autoencoders, which operate on jets in their "particle cloud" representations and can leverage the interdependencies among the particles within a jet, for such tasks. Additionally, we develop a differentiable approximation to the energy mover's distance via a graph neural network, which may subsequently be used as a reconstruction loss function for autoencoders.

arXiv comment5 pages, 2 figures. Accepted to the Machine Learning for the Physical Sciences workshop at NeurIPS 2021. arXiv admin note: text overlap with arXiv:2101.08320

References in corpus (19)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Deep Learning using Rectified Linear Units (ReLU)
• Fast Graph Representation Learning with PyTorch Geometric
• Interaction Networks for Learning about Objects, Relations and Physics
• ParticleNet: Jet Tagging via Particle Clouds
• Pileup Per Particle Identification
• Searching for New Physics with Deep Autoencoders
• Search for narrow and broad dijet resonances in proton-proton collisions at $\sqrt{s}=$ 13 TeV and constraints on dark matter mediators and other new particles
• Learning representations of irregular particle-detector geometry with distance-weighted graph networks
• The Metric Space of Collider Events
• Tag N' Train: A Technique to Train Improved Classifiers on Unlabeled Data
• On hypothesis testing, trials factor, hypertests and the BumpHunter
• Graph Neural Networks for Particle Reconstruction in High Energy Physics detectors
• Interaction networks for the identification of boosted $H\to b\overline{b}$ decays
• Bump Hunting in Latent Space
• Deep Learning as a Parton Shower
• FSPool: Learning Set Representations with Featurewise Sort Pooling
• Charged particle tracking via edge-classifying interaction networks
• Linearized Optimal Transport for Collider Events