DeepRICH: Learning Deeply Cherenkov Detectors
This work addresses the need for near real-time calibration and data analysis in particle physics experiments, though it appears incremental as it builds on existing deep learning techniques for a specific domain.
The authors tackled the problem of fast reconstruction for imaging Cherenkov detectors in particle physics by developing DeepRICH, a deep learning algorithm that bypasses low-level likelihood details, resulting in a sensitive improvement in computation time while maintaining similar reconstruction performance to established methods.
Imaging Cherenkov detectors are largely used for particle identification (PID) in nuclear and particle physics experiments, where developing fast reconstruction algorithms is becoming of paramount importance to allow for near real time calibration and data quality control, as well as to speed up offline analysis of large amount of data. In this paper we present DeepRICH, a novel deep learning algorithm for fast reconstruction which can be applied to different imaging Cherenkov detectors. The core of our architecture is a generative model which leverages on a custom Variational Auto-encoder (VAE) combined to Maximum Mean Discrepancy (MMD), with a Convolutional Neural Network (CNN) extracting features from the space of the latent variables for classification. A thorough comparison with the simulation/reconstruction package FastDIRC is discussed in the text. DeepRICH has the advantage to bypass low-level details needed to build a likelihood, allowing for a sensitive improvement in computation time at potentially the same reconstruction performance of other established reconstruction algorithms. In the conclusions, we address the implications and potentialities of this work, discussing possible future extensions and generalization.