From raw data to neutrino candidates: a neural-network pipeline for Baikal-GVD

TL;DR

This paper introduces a transformer-based neural network pipeline for Baikal-GVD that enhances neutrino event reconstruction, significantly speeds up processing, and improves accuracy over traditional methods.

Contribution

The authors develop a novel three-stage neural network pipeline with domain adaptation for real-time neutrino event classification in Baikal-GVD.

Findings

Orders-of-magnitude speedup over standard methods

Neural noise suppression outperforms algorithmic approaches

Effective domain adaptation improves simulation-data agreement

Abstract

We present a neural-network-based data processing pipeline for Baikal-GVD, designed to improve event reconstruction quality and accelerate neutrino candidates selection. The pipeline comprises three stages: fast suppression of extensive air shower events, suppression of noise optical modules activations, and extraction of high confidence neutrino candidates. All three networks employ a transformer architecture that exploits inter-hit correlations through the attention mechanism. Applied sequentially, the pipeline achieves orders-of-magnitude speedup over the standard reconstruction chain. Moreover, noise suppression neural network surpasses the accuracy of algorithmic noise suppression algorithms and provides estimate for time residuals of the signal hits, which is crucial for identification of track-like hits. We address the domain shift between Monte Carlo simulations and experimental data by incorporating a domain adaptation technique, demonstrating improved agreement between the two domains. The resulting framework enables near-real-time event classification, with direct applications to multi-messenger alert systems and diffuse neutrino flux measurements.