paper-with-me

Papers

Learning Efficient Representations of Neutrino Telescope Events

2024-10-17 · Felix J. Yu, Nicholas Kamp, Carlos A. Argüelles

Neutrino telescopes detect rare interactions of particles produced in some of the most extreme environments in the Universe. This is accomplished by instrumenting a cubic-kilometer volume of naturally occurring transparent medium with light sensors. Given their substantial size and the high frequency of background interactions, these telescopes amass an enormous quantity of large variance, high-dimensional data. These attributes create substantial challenges for analyzing and reconstructing interactions, particularly when utilizing machine learning (ML) techniques. In this paper, we present a novel approach, called om2vec, that employs transformer-based variational autoencoders to efficiently represent neutrino telescope events by learning compact and descriptive latent representations. We demonstrate that these latent representations offer enhanced flexibility and improved computational efficiency, thereby facilitating downstream tasks in data analysis.

📄 PDF Abstract BibTeX arXiv:2410.13148

Code (1)

felixyu7/om2vec 공식 구현 pytorch

Tasks

Computational EfficiencyDescriptive

Similar Papers 제목 키워드 기반

Enhancing Events in Neutrino Telescopes through Deep Learning-Driven Super-Resolution

2024-08-16 · Felix J. Yu, Nicholas Kamp, Carlos A. Argüelles

Recent discoveries by neutrino telescopes, such as the IceCube Neutrino Observatory, relied extensively on machine learning (ML) tools to infer physical quantities from the raw photon hits detected. Neutrino telescope re…

Super-Resolution

A Differentiable Neural Surrogate for Photon Propagation in Neutrino Telescopes

2026-09-04 · Felix J. Yu, Berthy T. Feng, Nicholas Kamp, Carlos A. Argüelles arxiv

Large-volume neutrino telescopes infer neutrino properties from Cherenkov light, but simulating the transport of billions of photons through highly scattering ice or water is computationally costly. We introduce candela,…

Trigger-Level Event Reconstruction for Neutrino Telescopes Using Sparse Submanifold Convolutional Neural Networks

2023-03-15 · Felix J. Yu, Jeffrey Lazar, Carlos A. Argüelles

Convolutional neural networks (CNNs) have seen extensive applications in scientific data analysis, including in neutrino telescopes. However, the data from these experiments present numerous challenges to CNNs, such as n…

GPU

GraphNeT: Graph neural networks for neutrino telescope event reconstruction

2022-10-21 · Andreas Søgaard, Rasmus F. Ørsøe, Leon Bozianu, Morten Holm 외

GraphNeT is an open-source python framework aimed at providing high quality, user friendly, end-to-end functionality to perform reconstruction tasks at neutrino telescopes using graph neural networks (GNNs). GraphNeT mak…

Question Answering

Reducing Simulation Dependence in Neutrino Telescopes with Masked Point Transformers

2025-10-02 · Felix J. Yu, Nicholas Kamp, Carlos A. Argüelles arxiv

Machine learning techniques in neutrino physics have traditionally relied on simulated data, which provides access to ground-truth labels. However, the accuracy of these simulations and the discrepancies between simulate…

Self-Supervised Learning