Machine Learning Uncertainties with Adversarial Neural Networks

TL;DR

This paper introduces a method using adversarial neural networks to incorporate known uncertainties into machine learning models, improving event classification and parameter fitting in particle physics.

Contribution

It presents a novel approach to include systematic and theoretical uncertainties during training of neural networks for particle physics applications.

Findings

Enhanced event classification accuracy.

Improved parameter fitting robustness.

Demonstrated effectiveness in Higgs boson production analysis.

Abstract

Machine Learning is a powerful tool to reveal and exploit correlations in a multi-dimensional parameter space. Making predictions from such correlations is a highly non-trivial task, in particular when the details of the underlying dynamics of a theoretical model are not fully understood. Using adversarial networks, we include a priori known sources of systematic and theoretical uncertainties during the training. This paves the way to a more reliable event classification on an event-by-event basis, as well as novel approaches to perform parameter fits of particle physics data. We demonstrate the benefits of the method explicitly in an example considering effective field theory extensions of Higgs boson production in association with jets.