Neutrino Physics and Machine Learning (NPML 2025)

Development of Machine Learning PID Methods in KamLAND-Zen Experiment

28 Oct 2025, 15:25

15m

Koshiba Hall (University of Tokyo)

Short talk (15min. + 5 min. Q/A) Experiments - 0nuBB

Speaker

Jun Nakane

Description

We developed machine learning methods for background rejection in the 0νββ decay search of the KamLAND-Zen experiment. Using CNN-based KamNet and Transformer-based ViViT for particle identification from PMT hit maps, we compared the rejection efficiency of both models. The results showed equivalent performance with high correlation (0.85-0.95) in output scores. Performance improvement through integrated models was limited.