End-to-end optimisation of HEP triggers

TL;DR

This paper introduces an end-to-end differentiable framework for optimizing high-energy physics trigger systems, significantly improving detection performance while maintaining interpretability and constraints.

Contribution

It formulates trigger design as a unified end-to-end optimization problem, integrating all stages into a single differentiable system for the first time.

Findings

2-4 times increase in true-positive rate at fixed false-positive rate

Preserves interpretability of physics objects

Maintains calibration constraints

Abstract

High-energy physics experiments face extreme data rates, requiring real-time trigger systems to reduce event throughput while preserving sensitivity to rare processes. Trigger systems are typically constructed as modular chains of sequentially optimised algorithms, including machine learning models. Each algorithm is optimised for a specific local objective with no guarantee of overall optimality. We instead formulate trigger design as a constrained end-to-end optimisation problem, treating all stages- including data encoding, denoising, clustering, and calibration- as components of a single differentiable system trained against a unified physics objective. The framework jointly optimises performance while incorporating physics and deployment constraints. We demonstrate this approach on a hardware multi-jet trigger inspired by the ATLAS High-Luminosity Large Hadron Collider design. Using Higgs boson pair production as a benchmark, we observe x2-4 improvement in true-positive rate at fixed false-positive rate, while preserving interpretable intermediate physics objects and monotonic calibration constraints. These results highlight end-to-end optimisation as a practical paradigm for next-generation real-time event selection systems.