Deep-learning-based low-energy trigger algorithms for the Hyper-Kamiokande experiment
Researchers have developed deep-learning algorithms to improve the detection of low-energy neutrino events in the Hyper-Kamiokande experiment. These new methods, including supervised classifiers and anomaly detection models, significantly outperform traditional trigger systems in identifying signals. The deep learning approaches achieve high identification efficiencies for events as low as 3 MeV, with inference times fast enough for real-time operation on GPUs. AI
IMPACT Enhances scientific discovery by improving data analysis in particle physics experiments.