I. Kharuk

I. Kharuk, G. Rubtsov, G. Safronov

Baikal-GVD is a large ($\sim$1 km$^3$) underwater neutrino telescope installed in the fresh waters of Lake Baikal. The deep lake water environment is pervaded by background light, which is detectable by Baikal-GVD's photosensors. We introduce a neural network for an efficient separation of these noise hits from the signal ones, stemming from the propagation of relativistic particles through the detector. The model has a U-net-like architecture and employs temporal (causal) structure of events. The neural network's metrics reach up to 99\% signal purity (precision) and 96\% survival efficiency (recall) on Monte-Carlo simulated dataset. We compare the developed method with the algorithmic approach to rejecting the noise and discuss other possible architectures of neural networks, including graph-based ones.

A. Matseiko, G. Plotnikov, I. Kharuk

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.

O. Kalashev, I. Kharuk, M. Kuznetsov et al.

We introduce a novel method for identifying the mass composition of ultra-high-energy cosmic rays using deep learning. The key idea of the method is to use a chain of two neural networks. The first network predicts the type of a primary particle for individual events, while the second infers the mass composition of an ensemble of events. We apply this method to the Monte-Carlo data for the Telescope Array Surface Detectors readings, on which it yields an unprecedented low error of 7% for 4-component approximation. We also discuss the problems of applying the developed method to the experimental data, and the way they can be resolved.