Tau identification for VBF H?tautau

1
Tau identification for VBF H?tautau

• Koji Nakamura (Tsukuba)
• Junichi Tanaka(Tokyo)
• Shoji Asai(Tokyo)

Contents

• Motivation
• Tau Identification focusing on low Pt Tau
• Conclusion and Plan

2
Motivation
Invariant mass distribution of VBF H?tautau
(MH120GeV) with fast simulation
VBF H?tau tau is promising channel for Higgs
discovery
Z?tt
But as we can see the statistics of the signal is
limited.
Signal
Pt distribution of tau jet (MH125GeV)
Our Pt threshold of tau jet is 40GeV
(SN-ATLAS-2003-024) But many signals are in the
lower Pt region.
Low Pt Tau jet identification and reconstruction
are very important to increase statistics.
? We have developed tau identification optimized
low Pt region
3
MC Samples
Signal
VBF Higgs?tau(l) tau(h) MH125GeV Version
9.0.4(Sim) 10.0.4(Rec)
Background
di-jet sample J1(17-35GeV), J2(35-70GeV),
J3(70-140GeV) Version 9.0.4(Sim) 10.0.4(Rec)
4
Tau identification

• Tau candidates are selected based on track
  (finding Isolate 1 track or 3 tracks)
  ATL-COM-PHYS-2004-057,080
• We also use cells information .

We also required isolation there is no more
Ptgt2GeV track in DRlt0.4.
Find Pt gt5 GeV tracks
only 1 track in DRlt0.2
3tracks in DRlt0.1 (Find two more Ptgt2GeV tracks)
Find a seed cell associated with this track, and
make Cluster (The seed is defined as a cell with
larger than 6s of Noise near the track)
reconstruction
identification
Narrowness cut is applied for identification (
I will talk in later slides)
5
Definition of Efficiency/Rejection

• We defined Tagging Efficiency
• Rejection factor ( using di-jet sample J1,J2,J3)

Matching condition nearest Mctruth Jet around
hard process
parton is in DRlt0.4
6
Cell-based narrowness
phi
eta
We use cell based narrowness definition not DR.
Tile Sampling 0 LArHEC Sampling 0
LArEM Sampling 2
If Seed Cell is not on LArEM
Seed Cell
7
Narrowness distribution after 1prong/3prong
selection
We can enhance the signal against the BG by
setting narrowness threshold. For example
0.7/0.6(1prong/3prong)
8
Tagging Efficiency
We plotted efficiency as the function of visible
Tau Pt. Efficiency was evaluated with various
narrowness threshold.
Our result (Track base)
Tau Bank (Cell base)
Tagging Efficiency
Tagging Efficiency
Visible tau Pt GeV
Visible tau Pt GeV
We have similar efficiency as the Cell based
result when narrowness threshold is set to
0.7/0.6 (1prong/3prong)
9
Rejection factor
Lgt2
Lgt4
Cell base
1/R
1pr0.7 3pr0.6
1pr0.9 3pr0.8
Track base
1/R
Mctruth Jet Pt GeV
Mctruth Jet Pt GeV
These figures show that
Our track based result is not so good as cell
base result. But
10
Track multiplicity in Jet
MCtruth
J3(70-140GeV) Sample
1.5 becomes 1pr/3pr candidate
We selected charged particles in Mctruth e,m,p,K,p
Mean 13.0
Rec Track (iPad)
But We also checked reconstructed track (iPad
and xKalman) , the number of tracks is too small !
10 becomes 1pr/3pr candidate
This difference gives a critical effect on our
tau identification !
Mean 6.3
of Tracks in 1 Jet
11
Conclusion

• We checked identification efficiency of hadronic
  tau from VBF H-gttau tau
• As the tau identification algorithm, we used
  track based reconstruction and cell narrowness
  identification.
• We have similar efficiency as the cell based
  result when narrowness threshold is set to
  0.7/0.6.

plan

• In order to improve rejection performance , we
  need to understand the reason of low track
  reconstruction efficiency in Jet.
• We will parameterize performance of our tau
  identification to apply to Atlfast. We estimate
  the contribution from Wjjj and other QCD
  background when we set the visible tau Pt
  threshold lower.

12
Back Up