🤖 AI Summary
To address the challenges of background suppression and low-accuracy particle semantic labeling in neutrino event reconstruction for Liquid Argon Time Projection Chambers (LArTPCs), this paper proposes a heterogeneous graph-based multi-plane joint modeling approach. It represents multi-view 2D energy depositions as a heterogeneous graph and designs a multi-head attention-driven message-passing mechanism that fuses 2D observations with 3D geometric consistency constraints. A lightweight, transferable graph neural network architecture is developed, enabling cross-detector deployment and multi-task extensibility. Experiments demonstrate a primary interaction identification accuracy of 98.0% and a particle-type labeling accuracy of 94.9%. Inference time per event is merely 0.12 seconds on CPU and drops to 0.005 seconds/event under GPU batch processing. This work achieves, for the first time, end-to-end LArTPC event reconstruction with high accuracy, high efficiency, and strong generalizability.
📝 Abstract
Liquid Argon Time Projection Chamber (LArTPC) detector technology offers a wealth of high-resolution information on particle interactions, and leveraging that information to its full potential requires sophisticated automated reconstruction techniques. This article describes NuGraph2, a Graph Neural Network (GNN) for low-level reconstruction of simulated neutrino interactions in a LArTPC detector. Simulated neutrino interactions in the MicroBooNE detector geometry are described as heterogeneous graphs, with energy depositions on each detector plane forming nodes on planar subgraphs. The network utilizes a multi-head attention message-passing mechanism to perform background filtering and semantic labelling on these graph nodes, identifying those associated with the primary physics interaction with 98.0% efficiency and labelling them according to particle type with 94.9% efficiency. The network operates directly on detector observables across multiple 2D representations, but utilizes a 3D-context-aware mechanism to encourage consistency between these representations. Model inference takes 0.12 s/event on a CPU, and 0.005 s/event batched on a GPU. This architecture is designed to be a general-purpose solution for particle reconstruction in neutrino physics, with the potential for deployment across a broad range of detector technologies, and offers a core convolution engine that can be leveraged for a variety of tasks beyond the two described in this article.