Speaker
Description
Phenomenological studies at particle colliders rely on high-fidelity simulation data. These simulations are typically produced using Monte Carlo event generators such as Pythia and Herwig. However, several components of the event-generation pipeline, most notably hadronization, are sensitive to nonperturbative quantum chromodynamics (QCD), where first-principles understanding is limited. As a result, traditional event generators depend on phenomenological hadronization models whose parameters are tuned to experimental data, potentially introducing systematic uncertainties and causing deviations from experimental measurements. Recent advances in classical machine learning have begun to explore data-driven approaches to hadronization using generative models, which offer greater flexibility than traditional parameterized models and may thus improve simulation accuracy. Because hadronization is fundamentally a quantum process, quantum machine learning may provide a particularly natural framework for modeling its underlying structure. Motivated by this possibility, we present a proof-of-concept study of generative quantum machine learning applied to hadronization.