Deep Learning-based Tagger for Boosted WW(*) Semi-leptonic Decays at ATLAS

14 Jul 2021, 16:45
15m
Track B (Zoom)

Track B

Zoom

talk Higgs & Electroweak Physics Higgs & Electroweak Physics

Speaker

Dewen Zhong (Univ. Illinois at Urbana Champaign (US))

Description

Identifying $𝑊𝑊^{(\ast)}\rightarrow\ell\nu qq$ from heavy particle decays at the LHC is an important but challenging problem due to overlapping lepton and jet signatures. We have developed a deep learning-based $𝑊𝑊^{(\ast)}$ tagger which learns from simulated calorimeter features to identify boosted $𝑊𝑊^{(\ast)}$ decays to semileptonic final states from $t\bar{t}$ and di-jet backgrounds in ATLAS. In this talk, we present the methods applied to the tagger development in the electron channel and some preliminary performance results on simulated ATLAS events at $\sqrt{s}=13$ TeV.

Are you are a member of the APS Division of Particles and Fields? Yes

Primary author

Dewen Zhong (Univ. Illinois at Urbana Champaign (US))

Presentation materials