Deep-learned Top Tagging with a Lorentz Layer
Top Cited Papers
Open Access
- 1 September 2018
- journal article
- research article
- Published by Stichting SciPost in SciPost Physics
- Vol. 5 (3), 028
- https://doi.org/10.21468/SciPostPhys.5.3.028
Abstract
We introduce a new and highly efficient tagger for hadronically decaying top quarks, based on a deep neural network working with Lorentz vectors and the Minkowski metric. With its novel machine learning setup and architecture it allows us to identify boosted top quarks not only from calorimeter towers, but also including tracking information. We show how the performance of our tagger compares with QCD-inspired and image-recognition approaches and find that it significantly increases the performance for strongly boosted top quarks.Funding Information
- Deutsche Forschungsgemeinschaft (GRK 1940)
- European Commission (PITN-GA-2012-315877)
This publication has 46 references indexed in Scilit:
- Boosted objects: a probe of beyond the standard model physicsThe European Physical Journal C, 2011
- Substructure of high-jets at the LHCPhysical Review D, 2009
- Top quark jets at the LHCPhysical Review D, 2009
- Top Tagging: A Method for Identifying Boosted Hadronically Decaying Top QuarksPhysical Review Letters, 2008
- Color-octet scalars at the CERN LHCPhysical Review D, 2008
- The anti-ktjet clustering algorithmJournal of High Energy Physics, 2008
- Using jet mass to discover vector quarks at the CERN LHCPhysical Review D, 2007
- t′ at the LHC: the physics of discoveryJournal of High Energy Physics, 2007
- Dispelling the N3 myth for the kt jet-finderPhysics Letters B, 2006
- scattering at the CERN LHCPhysical Review D, 2002