Particle Hit Clustering and Identification Using Point Set Transformers in Liquid Argon Time Projection Chambers

2025-04-11

AI Summary

In liquid-argon time-projection chambers, hit data are extremely sparse—most voxels carry zero energy—rendering conventional CNNs inefficient due to mandatory input densification, which incurs prohibitive memory and computational overhead. This work introduces the first application of point-set Transformers to high-energy physics sparse detector data, directly processing raw 3D spatial coordinates and energy deposits without voxelization. We propose three key innovations: sparse coordinate embedding, adaptive neighborhood attention, and a lightweight feature aggregation network. Compared to state-of-the-art sparse methods, our approach improves classification and segmentation accuracy by 14% and 22%, respectively, while reducing inference time by 80% and memory consumption by 66%. Against state-of-the-art CNNs, it achieves 86% and 71% accuracy gains in classification and segmentation, with 91% faster inference and 61% lower memory usage. Our method thus delivers dual breakthroughs: geometric fidelity preservation and a paradigm shift in efficient sparse-data representation.

Technology Category

• Machine Learning: Hardware-aware ML
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Security and Privacy: Data transparency and provenance
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Liquid argon time projection chambers are often used in neutrino physics and dark-matter searches because of their high spatial resolution. The images generated by these detectors are extremely sparse, as the energy values detected by most of the detector are equal to 0, meaning that despite their high resolution, most of the detector is unused in a particular interaction. Instead of representing all of the empty detections, the interaction is usually stored as a sparse matrix, a list of detection locations paired with their energy values. Traditional machine learning methods that have been applied to particle reconstruction such as convolutional neural networks (CNNs), however, cannot operate over data stored in this way and therefore must have the matrix fully instantiated as a dense matrix. Operating on dense matrices requires a lot of memory and computation time, in contrast to directly operating on the sparse matrix. We propose a machine learning model using a point set neural network that operates over a sparse matrix, greatly improving both processing speed and accuracy over methods that instantiate the dense matrix, as well as over other methods that operate over sparse matrices. Compared to competing state-of-the-art methods, our method improves classification performance by 14%, segmentation performance by more than 22%, while taking 80% less time and using 66% less memory. Compared to state-of-the-art CNN methods, our method improves classification performance by more than 86%, segmentation performance by more than 71%, while reducing runtime by 91% and reducing memory usage by 61%.

Problem

• Improving particle hit clustering in sparse detector data
• Enhancing processing speed and accuracy in particle reconstruction
• Reducing memory usage for sparse matrix operations

Innovation

• Uses point set neural networks
• Operates directly on sparse matrices
• Improves speed and accuracy significantly