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.
Description
Deep-learned Top Tagging using Lorentz Invariance and Nothing Else
%0 Journal Article
%1 butter2017deeplearned
%A Butter, Anja
%A Kasieczka, Gregor
%A Plehn, Tilman
%A Russell, Michael
%D 2017
%K hep-ex ml top
%T Deep-learned Top Tagging using Lorentz Invariance and Nothing Else
%U http://arxiv.org/abs/1707.08966
%X 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.
@article{butter2017deeplearned,
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.},
added-at = {2017-07-31T09:51:21.000+0200},
author = {Butter, Anja and Kasieczka, Gregor and Plehn, Tilman and Russell, Michael},
biburl = {https://www.bibsonomy.org/bibtex/207c0765f9b189ed0a1b2414749811f16/vindex10},
description = {Deep-learned Top Tagging using Lorentz Invariance and Nothing Else},
interhash = {dbbb35e831575f47c18332d2c80dff3a},
intrahash = {07c0765f9b189ed0a1b2414749811f16},
keywords = {hep-ex ml top},
note = {cite arxiv:1707.08966},
timestamp = {2017-07-31T09:51:21.000+0200},
title = {Deep-learned Top Tagging using Lorentz Invariance and Nothing Else},
url = {http://arxiv.org/abs/1707.08966},
year = 2017
}