HEP advanced tracking algorithms with cross-cutting applications (Project HEP.TrkX)
about
An HEP/ASCR DOE pilot project to evaluate and broaden the range of computational techniques and algorithms utilized in addressing HEP tracking challenges. Specifically the project demonstrated advanced pattern recognition techniques to track candidate formation.
HEP.TrkX is now concluded, and has been followed up by the DOE HEP
Exa.TrkX
project.
summary
HEP.TrkX github (public code, artifacts)
results
Paolo Calafiura (PI), Steve Farrell, Xiangyang Ju, Prabhat (LBNL)
Giuseppe Cerati, Lindsey Gray, Jim Kowalkowski, Panagiotis Spentzouris (co-PI), Aristeidis Tsaris (FNAL)
Maria Spiropulu (co-PI), Jean-Roch Vlimant, Alexander Zlokapa(Caltech)
people
aCTS
a Common Tracking Software
gitkit by Noemi Calace, Christian Gumpert, Julia Hrdinka & Andreas Salzburger
The Parallel Kalman Filter Project
NVIDIA
SuperMicro
collaborators, partners & toolkits
November 1-3 2016 1st HEP.TrKX workshop at FNAL
Indico agenda
October 16-17 2016 at LBNL hands-on working meeting after CHEP in SF
October 8,9,10 2016 during CPAD at Caltech
Indico agenda
meetings & workshops
Slack
PRIVATE
Development github
PRIVATE
Collaboration tools
Publications
Novel deep learning methods for track reconstruction
The HEP.TrkX Project: deep neural networks for HL-LHC online and offline tracking
Conference Contributions
HEP.TrkX Charged Particle Tracking using Graph Neural Networks
@
Connecting the Dots / Intelligent Trackers 2019 Workshop
HEP.TrkX: Novel deep learning methods for track reconstruction
@
Connecting the Dots 2018 Workshop
Particle Track Reconstruction with Deep Learning
@
Deep Learning for Physical Sciences NIPS 2017 Workshop
NIPS 2017 demo
(
poster
)
August 21-25 2017 ACAT
July 19-28 2017 Hammers and Nails
June 6-9 2017 ATLAS Machine Learning Workshop
Convolutional NNs for Tracking
DS@HEP Workshop at FNAL
HEP.TrkX
IML Machine Learning Workshop
The HEP.TrkX project: Deep Neural Networks for HEP Tracking
@
CTD/WIT 2017
ML LHC Tracking Challenge
@
CHEP 2016
reports, publications & challenges
project co-PI's
contact
powered by