Energy-weighted Message Passing: an infra-red and collinear safe graph neural network algorithm

TL;DR

This paper introduces an energy-weighted message-passing graph neural network that is infra-red and collinear safe, improving the identification of particle jets and outperforming existing methods.

Contribution

It develops a novel class of GNNs with IRC safety, enhancing interpretability and robustness, and generalizes Energy Flow Networks to graph structures.

Findings

Outperforms state-of-the-art Energy Flow Networks in jet classification

Ensures structural invariance of graphs in the IRC limit

Provides a general framework for IRC-safe graph construction

Abstract

Hadronic signals of new-physics origin at the Large Hadron Collider can remain hidden within the copiously produced hadronic jets. Unveiling such signatures require highly performant deep-learning algorithms. We construct a class of Graph Neural Networks (GNN) in the message-passing formalism that makes the network output infra-red and collinear (IRC) safe, an important criterion satisfied within perturbative QCD calculations. Including IRC safety of the network output as a requirement in the construction of the GNN improves its explainability and robustness against theoretical uncertainties in the data. We generalise Energy Flow Networks (EFN), an IRC safe deep-learning algorithm on a point cloud, defining energy weighted local and global readouts on GNNs. Applying the simplest of such networks to identify top quarks, W bosons and quark/gluon jets, we find that it outperforms state-of-the-art EFNs. Additionally, we obtain a general class of graph construction algorithms that give structurally invariant graphs in the IRC limit, a necessary criterion for the IRC safety of the GNN output.