GraphNeT 2.0 -- A Deep Learning Library for Neutrino Telescopes

TL;DR

GraphNeT 2.0 is an open-source, detector-agnostic deep learning library designed to enhance neutrino telescope data analysis through advanced neural network techniques, fostering collaboration across experiments.

Contribution

It introduces GraphNeT 2.0, a novel, versatile deep learning library tailored for neutrino telescopes, enabling cross-experimental collaboration and application of cutting-edge methods.

Findings

Improves prediction speed and accuracy in neutrino data analysis.

Provides a flexible, detector-agnostic framework for deep learning applications.

Facilitates collaboration across different neutrino experiments.

Abstract

Neutrino telescopes, an extension of traditional multiwavelength astronomy, provide a complementary view of the universe using neutrinos. Differences in detector geometry and detection medium mean that improvements to reconstruction techniques made at one experiment are not readily applicable to another. Recently, deep learning has been shown to improve prediction speed and accuracy and offer indifference to detector geometry and detection medium, providing a unique opportunity for collaboration. This work introduces GraphNeT 2.0, an open-source, detector-agnostic deep learning library for neutrino telescopes and related experiments. GraphNeT enables inter-experimental collaboration on the use and development of advanced methods based on major deep learning paradigms like transformers, normalizing flows, graph neural networks, and more.