HF Daily Papers
· Papers
Learning to Trigger: Reinforcement Learning at the Large Hadron Collider
High-throughput scientific facilities such as the Large Hadron Collider depend on real-time event filtering (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,