A ML-based direction estimate for ANTARES single-line events

ANTARES telescope is a neutrino detector residing 2.5 km under the Mediterranean Sea off the coast of France. It is designed to be used as a directional neutrino telescope to locate and observe neutrino flux from cosmic origins. It is composed of 12 vertical lines of photomultiplier tubes that detect the Cherenkov radiation of the muons produced by the interaction between neutrinos and water.

BBfit single-line reconstruction is, for low energy neutrinos, the most efficient algorithm in ANTARES. This energy range is highly relevant for studying Dark Matter and neutrino oscillations. However, BBfit only provides information about the Zenith angle of the neutrino direction.

Our approach is using Machine Learning techniques (Deep Neural Networks) to constrain not only the Zenith but also the Azimuth in single-line events by studying the distribution of photon hits at the photomultiplier tubes.