An Optimal Observable Machine for reinterpretable measurements in high-energy physics

TL;DR

This paper introduces the Optimal Observable Machine (OOM), a machine learning framework that constructs generator-level observables optimized for precise, robust parameter extraction in high-energy physics, exemplified by top quark studies.

Contribution

The paper presents a novel machine learning approach that explicitly incorporates detector effects and uncertainties to optimize observables for parameter measurement in particle physics.

Findings

Enhanced sensitivity to the pseudoscalar excess in top quark data

Robustness against detector effects and systematic uncertainties

Generation of interpretable, long-term usable observables

Abstract

A machine-learning-based framework for constructing generator-level observables optimized for parameter extraction in particle physics analyses is introduced, referred to as the Optimal Observable Machine (OOM). Unfoldable differential distributions are learned that maximize sensitivity to a parameter of interest while remaining robust against detector effects, systematic uncertainties, and biases introduced by the unfolding procedure. Detector response and systematic uncertainties are explicitly incorporated into the training through a likelihood-based loss function, enabling a direct optimization of the expected measurement precision while minimizing the bias from any assumption on the parameter of interest itself. The approach is demonstrated in an application to top quark physics, focusing on the measurement of a recently observed pseudoscalar excess at the top quark pair production threshold in dilepton final states. It is shown that a generator-level observable with enhanced sensitivity and long-term reinterpretability can be constructed using this method.