Neutrino Fingerprints: Image-Based Encodings of IceCube Events for CNN Direction Reconstruction
This provides an interpretable baseline for IceCube event reconstruction, but the performance is modest and the approach is incremental.
The authors introduced neutrino fingerprints, compact 72x72x3 images encoding pulse timing and charge statistics, enabling a ResNet18 CNN to achieve a mean angular error of 1.10 rad for neutrino direction reconstruction in IceCube, rivaling more complex architectures.
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.