Learning Tree Structures from Leaves For Particle Decay Reconstruction

Abstract

In this work, we present a neural approach to reconstructing rooted tree graphs describing hierarchical interactions, using a novel representation we term the Lowest Common Ancestor Generations (LCAG) matrix. This compact formulation is equivalent to the adjacency matrix, but enables learning a tree's structure from its leaves alone without the prior assumptions required if using the adjacency matrix directly. Employing the LCAG therefore enables the first end-to-end trainable solution which learns the hierarchical structure of varying tree sizes directly, using only the terminal tree leaves to do so. In the case of high-energy particle physics, a particle decay forms a hierarchical tree structure of which only the final products can be observed experimentally, and the large combinatorial space of possible trees makes an analytic solution intractable. We demonstrate the use of the LCAG as a target in the task of predicting simulated particle physics decay structures using both a Transformer encoder and a Neural Relational Inference encoder Graph Neural Network. With this approach, we are able to correctly predict the LCAG purely from leaf features for a maximum tree-depth of $8$ in $92.5\%$ of cases for trees up to $6$ leaves (including) and $59.7\%$ for trees up to $10$ in our simulated dataset.

arXiv comment14 pages, 6 figures, accepted in Machine Learning: Science and Technology

References in corpus (23)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• A Comprehensive Survey on Graph Neural Networks
• Particle-flow reconstruction and global event description with the CMS detector
• Hierarchical Graph Representation Learning with Differentiable Pooling
• The LHCb Trigger and its Performance in 2011
• ParticleNet: Jet Tagging via Particle Clouds
• Deep Learning and its Application to LHC Physics
• Jet reconstruction and performance using particle flow with the ATLAS Detector
• Jet Substructure at the Large Hadron Collider: A Review of Recent Advances in Theory and Machine Learning
• Efficient, reliable and fast high-level triggering using a bonsai boosted decision tree
• Neural Relational Inference for Interacting Systems
• Graph Neural Networks in Particle Physics
• JEDI-net: a jet identification algorithm based on interaction networks
• Unifying Graph Convolutional Neural Networks and Label Propagation
• The Full Event Interpretation -- An exclusive tagging algorithm for the Belle II experiment
• Graph Neural Networks for Particle Reconstruction in High Energy Physics detectors
• MLPF: Efficient machine-learned particle-flow reconstruction using graph neural networks
• The Tracking Machine Learning challenge : Accuracy phase
• Particle Transformer for Jet Tagging
• Graph Neural Networks for Particle Tracking and Reconstruction
• Shared Data and Algorithms for Deep Learning in Fundamental Physics
• Secondary Vertex Finding in Jets with Neural Networks
• Hierarchical Message-Passing Graph Neural Networks

Cited by in corpus (5)

• Improved particle-flow event reconstruction with scalable neural networks for current and future particle detectors
• Parnassus: An Automated Approach to Accurate, Precise, and Fast Detector Simulation and Reconstruction
• Scalable Multi-Task Learning for Particle Collision Event Reconstruction with Heterogeneous Graph Neural Networks
• PASCL: Supervised Contrastive Learning with Perturbative Augmentation for Particle Decay Reconstruction
• Graph-based Full Event Interpretation: a graph neural network for event reconstruction in Belle II