Learning to Trigger: Reinforcement Learning at the Large Hadron Collider

2026-06-22

AI Summary

This study addresses the limitations of static trigger thresholds in high-energy physics experiments, which fail to adapt to dynamic variations in detector conditions, pileup effects, and background composition, leading to signal inefficiency and resource waste. The authors formulate threshold optimization as a sequential decision-making problem and present the first implementation of reinforcement learning–based trigger control on real Large Hadron Collider (LHC) data. They introduce two novel algorithms, GFPO-F and GFPO-FR, which explicitly enforce background rate constraints during training. Leveraging both Monte Carlo simulations and real CMS data, the approach enables deployment without fine-tuning. Experimental results demonstrate significant improvements: within the allowed tolerance band, the fraction of time spent increases by up to 56% for the H_T observable and 28% for AD, with an accumulated gain in signal efficiency of up to 2%.

Abstract

High-throughput scientific facilities such as the Large Hadron Collider depend on real-time event filtering (\textit{triggering}) under tight constraints on bandwidth, latency, and storage. In practice, trigger menus are largely static and hand-tuned and can become suboptimal as detector conditions, pileup, and background composition drift over time. We cast online threshold tuning as a sequential decision-making problem: a reinforcement learning agent ingests streaming summaries of recent rates and signal-sensitive features and updates trigger thresholds to maximize signal efficiency while tracking a target background rate within a tolerance band. We adapt Group-Filtered Policy Optimization (GFPO) to streaming control and introduce two variants (GFPO-F, GFPO-FR) that enforce background rate feasibility during training. On a benchmark that emulates realistic collider operation, we study two representative triggers: a total transverse energy ($H_{T}$) trigger sensitive to pileup variation, and an anomaly-detection (AD) trigger based on reconstruction loss for rare or non-standard signatures. On Monte Carlo streams, our agent increases the fraction of in-tolerance time intervals by 48\% ($H_T$) and 28\% (AD), with a cumulative gain of up to 2\% in signal efficiency on those in-tolerance intervals. Transferring from simulation to \emph{real} collision data (CMS Run 283408), the same agent, without fine-tuning, achieves a 56\% ($H_T$) and 28\% (AD) in-tolerance improvement over baselines, with further signal-efficiency gain on both triggers. To our knowledge, this is the \emph{first} demonstration of RL-based trigger control on real Large Hadron Collider collision data. Code is available at https://github.com/Zixind/GFPO\_LHC.

Problem

• triggering
• Large Hadron Collider
• real-time event filtering
• background rate control
• signal efficiency

Innovation

• Reinforcement Learning
• Trigger System
• Large Hadron Collider
• GFPO
• Real-time Control