Neutrino Fingerprints: Image-Based Encodings of IceCube Events for CNN Direction Reconstruction

2026-06-01

AI Summary

This study addresses the challenge of reconstructing the incident direction of neutrino events in the IceCube detector by proposing a novel representation of sparse, irregular photomultiplier pulse data as dense 72×72×3 “neutrino fingerprint” images. In this encoding, each pixel corresponds to a photomultiplier tube position, with the three channels representing pulse arrival time and charge-related statistics. Leveraging this structured image format, the authors employ a ResNet18 convolutional neural network to process raw detector data end-to-end for direction reconstruction. Evaluated on 140 million simulated events, the method achieves a mean angular error of 1.10 radians, matching the performance of more complex models while offering greater computational efficiency and interpretability. This approach establishes a general and effective deep learning baseline for handling sparse data in high-energy physics experiments.

Abstract

Reconstructing the direction of incoming neutrinos in the IceCube Neutrino Observatory is an important problem in astrophysics. The public IceCube--Neutrinos in Deep Ice Kaggle competition provided 140 million simulated events to benchmark reconstruction techniques. To address this challenge from a novel perspective we introduce neutrino fingerprints compact $72 \times 72 \times 3$ images in which each pixel represents a single detector, with pulse timing and charge statistics encoded as color channels. This representation transforms sparse, irregular pulse data into dense images suitable for convolutional processing. Our ResNet18 model achieves a mean angular error of $1.10$ rad, indicating that convolutional networks trained on fingerprints rival more complex architectures while offering an effective, interpretable baseline for IceCube event reconstruction.

Problem

• neutrino direction reconstruction
• IceCube
• astrophysics
• event reconstruction

Innovation

• neutrino fingerprints
• image-based encoding
• CNN
• direction reconstruction
• IceCube