English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The complex events observed at the NOvA long-baseline neutrino oscillationexperiment contain vital information for understanding the most elusiveparticles in the standard model. The NOvA detectors observe interactions ofneutrinos from the NuMI beam at Fermilab. Associating the particles produced inthese interaction events to their source particles, a process known asreconstruction, is critical for accurately measuring key parameters of thestandard model. Events may contain several particles, each producing sparsehigh-dimensional spatial observations, and current methods are limited toevaluating individual particles. To accurately label these numerous,high-dimensional observations, we present a novel neural network architecturethat combines the spatial learning enabled by convolutions with the contextuallearning enabled by attention. This joint approach, TransformerCVN,simultaneously classifies each event and reconstructs every individualparticle's identity. TransformerCVN classifies events with 90\% accuracy andimproves the reconstruction of individual particles by 6\% over baselinemethods which lack the integrated architecture of TransformerCVN. In addition,this architecture enables us to perform several interpretability studies whichprovide insights into the network's predictions and show that TransformerCVNdiscovers several fundamental principles that stem from the standard model.

Interpretable Joint Event-Particle Reconstruction for Neutrino Physics at NOvA with Sparse CNNs and Transformers