Unsupervised Learning for Identifying Events in Active Target Experiments

TL;DR

This paper demonstrates that unsupervised machine learning, particularly VGG16-based clustering, effectively groups similar particle events in active target experiments, enhancing data analysis efficiency.

Contribution

It introduces the use of unsupervised clustering with deep neural network features for event separation in active target detectors, showing near-perfect clustering in simulated data and high purity in real data.

Findings

VGG16 + k-means achieves near-perfect clustering in simulated data.

High purity clusters of proton events found in real experimental data.

Autoencoder-based clustering shows strong but variable performance.

Abstract

This article presents novel applications of unsupervised machine learning methods to the problem of event separation in an active target detector, the Active-Target Time Projection Chamber (AT-TPC). The overarching goal is to group similar events in the early stages of the data analysis, thereby improving efficiency by limiting the computationally expensive processing of unnecessary events. The application of unsupervised clustering algorithms to the analysis of two-dimensional projections of particle tracks from a resonant proton scattering experiment on $^{46}$Ar is introduced. We explore the performance of autoencoder neural networks and a pre-trained VGG16 convolutional neural network. We study clustering performance on both data from a simulated $^{46}$Ar experiment, and real events from the AT-TPC detector. We find that a $k$-means algorithm applied to simulated data in the VGG16 latent space forms almost perfect clusters. Additionally, the VGG16+$k$-means approach finds high purity clusters of proton events for real experimental data. We also explore the application of clustering the latent space of autoencoder neural networks for event separation. While these networks show strong performance, they suffer from high variability in their results.