Generative Models for Simulation of KamLAND-Zen
This work addresses the critical challenge of simulating extremely rare events for neutrino physics experiments, though it appears incremental as it supplements traditional Monte Carlo methods with machine learning.
The paper tackles the need for accurate simulations of rare neutrinoless double beta decay events in detectors like KamLAND, using generative models to produce simulation data without a predefined physics model, achieving recovery of low-level features and interpolation capabilities.
The next generation of searches for neutrinoless double beta decay (0ν\b{eta}\b{eta}) are poised to answer deep questions on the nature of neutrinos and the source of the Universe's matter-antimatter asymmetry. They will be looking for event rates of less than one event per ton of instrumented isotope per year. To claim discovery, accurate and efficient simulations of detector events that mimic 0ν\b{eta}\b{eta} is critical. Traditional Monte Carlo (MC) simulations can be supplemented by machine-learning-based generative models. In this work, we describe the performance of generative models designed for monolithic liquid scintillator detectors like KamLAND to produce highly accurate simulation data without a predefined physics model. We demonstrate its ability to recover low-level features and perform interpolation. In the future, the results of these generative models can be used to improve event classification and background rejection by providing high-quality abundant generated data.