Rejecting noise in Baikal-GVD data with neural networks
This work addresses noise rejection for neutrino detection in underwater telescopes, representing an incremental improvement over existing algorithmic methods.
The paper tackles the problem of separating noise hits from signal hits in the Baikal-GVD neutrino telescope data, achieving up to 99% signal purity and 96% survival efficiency using a neural network with a U-net-like architecture.
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.