A neural network for beam background decomposition in Belle II at SuperKEKB

TL;DR

This paper introduces BGNet, a neural network model that predicts and interprets the contributions of various physical background sources to detector hit rates in the Belle II experiment at SuperKEKB, improving understanding of background dynamics.

Contribution

The paper presents a novel neural network approach, BGNet, for decomposing and predicting background contributions in collider detector data, with interpretability features.

Findings

BGNet accurately predicts detector background rates across different run periods.

Feature attribution helps identify sources of background fluctuations.

Model enhances understanding of background dynamics in collider experiments.

Abstract

We describe a neural network for predicting the background hit rate in the Belle II detector produced by the SuperKEKB electron-positron collider. The neural network, BGNet, learns to predict the individual contributions of different physical background sources, such as beam-gas scattering or continuous top-up injections into the collider, to Belle II sub-detector rates. The samples for learning are archived 1 Hz time series of diagnostic variables from the SuperKEKB collider subsystems and measured hit rates of Belle II used as regression targets. We test the learned model by predicting detector hit rates on archived data from different run periods not used during training. We show that a feature attribution method can help interpret the source of changes in the background level over time.