Learning to Simulate High Energy Particle Collisions from Unlabeled Data

TL;DR

OTUS is an unsupervised machine learning-based simulator that efficiently predicts experimental data from theoretical models, potentially replacing costly traditional simulations in particle physics and other fields.

Contribution

Introduces OTUS, a novel unsupervised autoencoder approach that directly maps theoretical models to experimental data without relying on existing simulation data.

Findings

Successfully applied to Z-boson decay data

Accurately predicts top-quark decay distributions

Offers a faster alternative to traditional simulations

Abstract

In many scientific fields which rely on statistical inference, simulations are often used to map from theoretical models to experimental data, allowing scientists to test model predictions against experimental results. Experimental data is often reconstructed from indirect measurements causing the aggregate transformation from theoretical models to experimental data to be poorly-described analytically. Instead, numerical simulations are used at great computational cost. We introduce Optimal-Transport-based Unfolding and Simulation (OTUS), a fast simulator based on unsupervised machine-learning that is capable of predicting experimental data from theoretical models. Without the aid of current simulation information, OTUS trains a probabilistic autoencoder to transform directly between theoretical models and experimental data. Identifying the probabilistic autoencoder's latent space with the space of theoretical models causes the decoder network to become a fast, predictive simulator with the potential to replace current, computationally-costly simulators. Here, we provide proof-of-principle results on two particle physics examples, $Z$-boson and top-quark decays, but stress that OTUS can be widely applied to other fields.