skip to main content
DOE PAGES title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: DeepRICH: learning deeply Cherenkov detectors

Abstract

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.

Authors:
ORCiD logo; ORCiD logo
Publication Date:
Research Org.:
Thomas Jefferson National Accelerator Facility (TJNAF), Newport News, VA (United States); Massachusetts Inst. of Technology (MIT), Cambridge, MA (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Nuclear Physics (NP)
OSTI Identifier:
1616342
Alternate Identifier(s):
OSTI ID: 1616343; OSTI ID: 1616671; OSTI ID: 1623320
Report Number(s):
JLAB-PHY-20-3179; DOE/OR/23177-4965
Journal ID: ISSN 2632-2153
Grant/Contract Number:  
SC0019999; FG02-94ER40818
Resource Type:
Published Article
Journal Name:
Machine Learning: Science and Technology
Additional Journal Information:
Journal Name: Machine Learning: Science and Technology Journal Volume: 1 Journal Issue: 1; Journal ID: ISSN 2632-2153
Publisher:
IOP Publishing
Country of Publication:
United Kingdom
Language:
English
Subject:
46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND TECHNOLOGY; Cherenkov detectors; deep learning; particle identification; near real time algorithm; experimental particle physics

Citation Formats

Fanelli, Cristiano, and Pomponi, Jary. DeepRICH: learning deeply Cherenkov detectors. United Kingdom: N. p., 2020. Web. doi:10.1088/2632-2153/ab845a.
Fanelli, Cristiano, & Pomponi, Jary. DeepRICH: learning deeply Cherenkov detectors. United Kingdom. doi:https://doi.org/10.1088/2632-2153/ab845a
Fanelli, Cristiano, and Pomponi, Jary. Sun . "DeepRICH: learning deeply Cherenkov detectors". United Kingdom. doi:https://doi.org/10.1088/2632-2153/ab845a.
@article{osti_1616342,
title = {DeepRICH: learning deeply Cherenkov detectors},
author = {Fanelli, Cristiano and Pomponi, Jary},
abstractNote = {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.},
doi = {10.1088/2632-2153/ab845a},
journal = {Machine Learning: Science and Technology},
number = 1,
volume = 1,
place = {United Kingdom},
year = {2020},
month = {3}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record
DOI: https://doi.org/10.1088/2632-2153/ab845a

Save / Share:

Works referenced in this record:

The GlueX DIRC project
journal, July 2016


TOP counter for particle identification at the Belle II experiment
journal, December 2014

  • Inami, Kenji
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 766
  • DOI: 10.1016/j.nima.2014.07.006

TORCH: Time of flight identification with Cherenkov radiation
journal, May 2011

  • Charles, M. J.; Forty, R.
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 639, Issue 1
  • DOI: 10.1016/j.nima.2010.09.021

Machine learning for imaging Cherenkov detectors
journal, February 2020


Simulation and reconstruction of the PANDA Barrel DIRC
journal, December 2014

  • Dzhygadlo, R.; Gerhardt, A.; Go¨tzen, K.
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 766
  • DOI: 10.1016/j.nima.2014.04.073

Hybrid mesons from quenched QCD
journal, May 1997


The DIRC particle identification system for the BaBar experiment
journal, February 2005

  • Adam, I.; Aleksan, R.; Amerman, L.
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 538, Issue 1-3
  • DOI: 10.1016/j.nima.2004.08.129

Kernel estimation in high-energy physics
journal, May 2001


Hybrid mesons
journal, May 2015


Progress on development of the new FDIRC PID detector
journal, August 2013

  • Va'vra, J.; Arnaud, N.; Barnyakov, A. Yu.
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 718
  • DOI: 10.1016/j.nima.2012.11.112

Machine Learning in High Energy Physics Community White Paper
journal, September 2018


Design and performance of the focusing DIRC detector
journal, March 2015

  • Dey, B.; Borsato, M.; Arnaud, N.
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 775
  • DOI: 10.1016/j.nima.2014.11.120

Cherenkov detectors fast simulation using neural networks
journal, February 2020

  • Derkach, Denis; Kazeev, Nikita; Ratnikov, Fedor
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 952
  • DOI: 10.1016/j.nima.2019.01.031

Likelihood analysis of patterns in a time-of-propagation (TOP) counter
journal, September 2008

  • Starič, M.; Inami, K.; Križan, P.
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 595, Issue 1
  • DOI: 10.1016/j.nima.2008.07.018

Photo-ionization and Cherenkov ring imaging
journal, May 1977


High-performance DIRC detector for the future Electron Ion Collider experiment
journal, July 2016


FastDIRC: a fast Monte Carlo and reconstruction algorithm for DIRC detectors
journal, October 2016


Status of the GlueX DIRC
journal, February 2020

  • Patsyuk, M.; Ali, A.; Barbosa, F.
  • Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 952
  • DOI: 10.1016/j.nima.2018.12.071

The PANDA Barrel DIRC
journal, May 2016