Physics-Informed Graph Neural Networks for Transverse Momentum Estimation in CMS Trigger Systems

2025-07-25

AI Summary

In high-energy physics real-time triggering, existing models suffer from degraded $p_T$ estimation accuracy under high pileup and lack explicit physical modeling capabilities. To address this, we propose a physics-informed graph neural network framework. Our method innovatively designs four physics-aware graph structures—station-level, feature-level, bending-angle, and pseudorapidity graphs—incorporates a gated attention message-passing mechanism, and introduces a customized loss function integrating prior knowledge of the $p_T$ distribution. Evaluated on the CMS trigger dataset, our approach significantly improves robustness and inference efficiency: the station-level EdgeConv variant achieves a state-of-the-art MAE of 0.8525 with 55% fewer parameters; further optimization via $eta$-centered graph construction enhances the accuracy–efficiency trade-off. This work establishes a new paradigm for physics-guided machine learning in resource-constrained environments.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Graphical Models
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems

Abstract

Real-time particle transverse momentum ($p_T$) estimation in high-energy physics demands algorithms that are both efficient and accurate under strict hardware constraints. Static machine learning models degrade under high pileup and lack physics-aware optimization, while generic graph neural networks (GNNs) often neglect domain structure critical for robust $p_T$ regression. We propose a physics-informed GNN framework that systematically encodes detector geometry and physical observables through four distinct graph construction strategies that systematically encode detector geometry and physical observables: station-as-node, feature-as-node, bending angle-centric, and pseudorapidity ($η$)-centric representations. This framework integrates these tailored graph structures with a novel Message Passing Layer (MPL), featuring intra-message attention and gated updates, and domain-specific loss functions incorporating $p_{T}$-distribution priors. Our co-design methodology yields superior accuracy-efficiency trade-offs compared to existing baselines. Extensive experiments on the CMS Trigger Dataset validate the approach: a station-informed EdgeConv model achieves a state-of-the-art MAE of 0.8525 with $ge55%$ fewer parameters than deep learning baselines, especially TabNet, while an $η$-centric MPL configuration also demonstrates improved accuracy with comparable efficiency. These results establish the promise of physics-guided GNNs for deployment in resource-constrained trigger systems.

Problem

• Real-time particle transverse momentum estimation under hardware constraints
• Static models degrade in high pileup and lack physics-aware optimization
• Generic GNNs neglect domain structure for robust pT regression

Innovation

• Physics-informed GNN with four graph strategies
• Novel Message Passing Layer with attention
• Domain-specific loss functions with pT priors