20-24 October 2025
Mexico/General timezone
Home > Timetable > Session details > Contribution details

Contribution Poster Presentation

High-Energy Event Classification for Cherenkov Detectors Using Vision Transformers

Speakers

Primary authors

Co-authors

Summary

We present research on classifying high-energy events for simulated Water Cherenkov detectors. The goal is to evaluate the performance of supervised learning models based on the Vision Transformer (ViT) architecture for distinguishing between two types of high-energy events: electron-neutrino and gamma events. The Vision Transformer achieved up to 85% accuracy and a maximum ROC-AUC of 0.95, significantly outperforming traditional convolutional and resnet neural network architectures.

correo electrónico

felipe.orozco@udg.mx

Speaker

Felipe de Jesus Orozco Luna