CapsNets Continuing the Convolutional Quest
High Energy Physics - Phenomenology
2020-02-12 v3
Abstract
Capsule networks are ideal tools to combine event-level and subjet information at the LHC. After benchmarking our capsule network against standard convolutional networks, we show how multi-class capsules extract a resonance decaying to top quarks from both, QCD di-jet and the top continuum backgrounds. We then show how its results can be easily interpreted. Finally, we use associated top-Higgs production to demonstrate that capsule networks can work on overlaying images to go beyond calorimeter information.
Cite
@article{arxiv.1906.11265,
title = {CapsNets Continuing the Convolutional Quest},
author = {Sascha Diefenbacher and Hermann Frost and Gregor Kasieczka and Tilman Plehn and Jennifer M. Thompson},
journal= {arXiv preprint arXiv:1906.11265},
year = {2020}
}
Comments
Minor changes to text