Machine Learning for Particle Flow Reconstruction at CMS

TL;DR

This paper presents a machine learning-based particle flow algorithm for CMS that improves event reconstruction by integrating graph neural networks and explores its implementation on heterogeneous computing platforms like GPUs.

Contribution

It introduces a novel machine learning approach using graph neural networks for particle flow reconstruction and evaluates its performance and scalability within CMS software.

Findings

Enhanced reconstruction accuracy for jets and missing transverse energy.

Linear scaling of runtime and memory with input size.

Feasibility of deploying on heterogeneous platforms like GPUs.

Abstract

We provide details on the implementation of a machine-learning based particle flow algorithm for CMS. The standard particle flow algorithm reconstructs stable particles based on calorimeter clusters and tracks to provide a global event reconstruction that exploits the combined information of multiple detector subsystems, leading to strong improvements for quantities such as jets and missing transverse energy. We have studied a possible evolution of particle flow towards heterogeneous computing platforms such as GPUs using a graph neural network. The machine-learned PF model reconstructs particle candidates based on the full list of tracks and calorimeter clusters in the event. For validation, we determine the physics performance directly in the CMS software framework when the proposed algorithm is interfaced with the offline reconstruction of jets and missing transverse energy. We also report the computational performance of the algorithm, which scales approximately linearly in runtime and memory usage with the input size.