Sparse Methods for Vector Embeddings of TPC Data

TL;DR

This paper demonstrates that sparse convolutional networks can effectively generate meaningful vector embeddings of TPC data, with pre-training enhancing the quality and showing potential for cross-detector applications.

Contribution

It introduces the use of sparse ResNet architectures for TPC data representation learning, highlighting their effectiveness even with random weights and benefits from pre-training.

Findings

Sparse ResNet provides useful event embeddings even untrained.

Pre-training improves embedding quality.

Cross-detector embedding transfer shows promise.

Abstract

Time Projection Chambers (TPCs) are versatile detectors that reconstruct charged-particle tracks in an ionizing medium, enabling sensitive measurements across a wide range of nuclear physics experiments. We explore sparse convolutional networks for representation learning on TPC data, finding that a sparse ResNet architecture, even with randomly set weights, provides useful structured vector embeddings of events. Pre-training this architecture on a simple physics-motivated binary classification task further improves the embedding quality. Using data from the GAseous Detector with GErmanium Tagging (GADGET) II TPC, a detector optimized for measuring low-energy $\beta$-delayed particle decays, we represent raw pad-level signals as sparse tensors, train Minkowski Engine ResNet models, and probe the resulting event-level embeddings which reveal rich event structure. As a cross-detector test, we embed data from the Active-Target TPC (AT-TPC) -- a detector designed for nuclear reaction studies in inverse kinematics -- using the same encoder. We find that even an untrained sparse ResNet model provides useful embeddings of AT-TPC data, and we observe improvements when the model is trained on GADGET data. Together, these results highlight the potential of sparse convolutional techniques as a general tool for representation learning in diverse TPC experiments.