Improved particle-flow event reconstruction with scalable neural networks for current and future particle detectors

TL;DR

This paper presents scalable neural network models, including graph neural networks and transformers, for particle-flow event reconstruction in high-granularity detectors, achieving up to 50% improvement in jet momentum resolution and hardware portability.

Contribution

The study introduces scalable machine learning models for particle reconstruction that outperform traditional algorithms and are adaptable across different hardware platforms.

Findings

Graph neural network improves jet transverse momentum resolution by up to 50%.

Hyperparameter tuning enhances model performance significantly.

Models are portable across Nvidia, AMD, and Habana hardware.

Abstract

Efficient and accurate algorithms are necessary to reconstruct particles in the highly granular detectors anticipated at the High-Luminosity Large Hadron Collider and the Future Circular Collider. We study scalable machine learning models for event reconstruction in electron-positron collisions based on a full detector simulation. Particle-flow reconstruction can be formulated as a supervised learning task using tracks and calorimeter clusters. We compare a graph neural network and kernel-based transformer and demonstrate that we can avoid quadratic operations while achieving realistic reconstruction. We show that hyperparameter tuning significantly improves the performance of the models. The best graph neural network model shows improvement in the jet transverse momentum resolution by up to 50% compared to the rule-based algorithm. The resulting model is portable across Nvidia, AMD and Habana hardware. Accurate and fast machine-learning based reconstruction can significantly improve future measurements at colliders.