Energy Flow Networks: Deep Sets for Particle Jets

TL;DR

This paper introduces Energy Flow Networks, a set-based deep learning approach for particle physics that respects physical safety constraints and improves event classification, demonstrated on quark vs gluon jet discrimination.

Contribution

The paper adapts the Deep Sets framework to particle physics, creating Energy Flow Networks that are infrared and collinear safe, and introduces Particle Flow Networks for more general energy dependence.

Findings

Achieved comparable or better performance than existing methods in quark-gluon jet discrimination.

Unified event representations with detector images and radiation moments.

Provided interpretable visualizations of learned event features.

Abstract

A key question for machine learning approaches in particle physics is how to best represent and learn from collider events. As an event is intrinsically a variable-length unordered set of particles, we build upon recent machine learning efforts to learn directly from sets of features or "point clouds". Adapting and specializing the "Deep Sets" framework to particle physics, we introduce Energy Flow Networks, which respect infrared and collinear safety by construction. We also develop Particle Flow Networks, which allow for general energy dependence and the inclusion of additional particle-level information such as charge and flavor. These networks feature a per-particle internal (latent) representation, and summing over all particles yields an overall event-level latent representation. We show how this latent space decomposition unifies existing event representations based on detector images and radiation moments. To demonstrate the power and simplicity of this set-based approach, we apply these networks to the collider task of discriminating quark jets from gluon jets, finding similar or improved performance compared to existing methods. We also show how the learned event representation can be directly visualized, providing insight into the inner workings of the model. These architectures lend themselves to efficiently processing and analyzing events for a wide variety of tasks at the Large Hadron Collider. Implementations and examples of our architectures are available online in our EnergyFlow package.