Research output: Contribution to journal › Article › peer-review
Rejecting noise in Baikal-GVD data with neural networks. / Kharuk, I.; Rubtsov, G.; Safronov, G.
In: Journal of Instrumentation, Vol. 18, No. 9, P09026, 01.09.2023.Research output: Contribution to journal › Article › peer-review
}
TY - JOUR
T1 - Rejecting noise in Baikal-GVD data with neural networks
AU - Kharuk, I.
AU - Rubtsov, G.
AU - Safronov, G.
N1 - Публикация для корректировки.
PY - 2023/9/1
Y1 - 2023/9/1
N2 - Baikal-GVD is a large (∼ 1 km3) 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, stemmng 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.
AB - Baikal-GVD is a large (∼ 1 km3) 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, stemmng 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.
KW - Analysis and statistical methods
KW - Data Processing
KW - Data analysis
KW - Pattern recognition, cluster finding, calibration and fitting methods
UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-85173173386&origin=inward&txGid=17d745c111d13f516377150141a63f58
UR - https://www.mendeley.com/catalogue/2213e847-201c-3c05-b563-407cb6d5b618/
U2 - 10.1088/1748-0221/18/09/P09026
DO - 10.1088/1748-0221/18/09/P09026
M3 - Article
VL - 18
JO - Journal of Instrumentation
JF - Journal of Instrumentation
SN - 1748-0221
IS - 9
M1 - P09026
ER -
ID: 59280488