Finding Neutrinos: ML Aiding in the Hunt for the Ghost Particle

Neutrinos' weak interactions make them a notoriously difficult particle to detect. With dedicated technological advancements and efforts, physicists have begun to better understand these leptons' role in the Standard Model of particle physics and explore their contribution to some major questions in the universe. Since the beginning of the 21st century, neutrino physicists have utilized basic machine learning (ML) methods to aid in their analyses. These early ML models often relied on highly pre-processed inputs that originated from traditional reconstruction methods and calculations. Neutrino experimentalists have since expanded their machine learning methods to leverage information earlier in the data pipeline and optimized their methods for specific detectors or data features. Furthermore, they have been exploring novel machine learning architectures and integrating important features, such as interpretability. This talk will highlight machine learning advancements in neutrino physics, both that have led to important physics results and that have the potential to progress the field.