An interpretable unsupervised representation learning for high precision measurement in particle physics

TL;DR

This paper introduces HistoAE, an unsupervised neural network with a histogram-based loss that learns interpretable, physically-structured representations for particle charge and position, achieving high-precision measurements and enabling detector simulations.

Contribution

The paper presents HistoAE, a novel unsupervised autoencoder with a histogram-based loss that enforces physical interpretability in the latent space for particle physics applications.

Findings

Achieves charge resolution of 0.25 e and position resolution of 3 μm.

Latent space corresponds to particle charge and impact position.

Comparable to conventional measurement methods.

Abstract

Unsupervised learning has been widely applied to various tasks in particle physics. However, existing models lack precise control over their learned representations, limiting physical interpretability and hindering their use for accurate measurements. We propose the Histogram AutoEncoder (HistoAE), an unsupervised representation learning network featuring a custom histogram-based loss that enforces a physically structured latent space. Applied to silicon microstrip detectors, HistoAE learns an interpretable two-dimensional latent space corresponding to the particle's charge and impact position. After simple post-processing, it achieves a charge resolution of $0.25\,e$ and a position resolution of $3\,\mu\mathrm{m}$ on beam-test data, comparable to the conventional approach. These results demonstrate that unsupervised deep learning models can enable physically meaningful and quantitatively precise measurements. Moreover, the generative capacity of HistoAE enables straightforward extensions to fast detector simulations.