Collider physics II: Jets 1 Tuesday, September 11, 12 Two aspects - - PowerPoint PPT Presentation

▶
collider physics ii jets
SMART_READER_LITE
LIVE PREVIEW

Collider physics II: Jets 1 Tuesday, September 11, 12 Two aspects - - PowerPoint PPT Presentation

Nov 17, 2022 127 likes •919 views

Collider physics II: Jets 1 Tuesday, September 11, 12 Two aspects of new developments Better QCD jet. - Smarter jet algorithm. - Noise suppression with jet grooming. Jet substructure. - Boosted top. - Higgs. Boston Jet Workshop:

slide-1
SLIDE 1

Collider physics II: Jets

1

Tuesday, September 11, 12

slide-2
SLIDE 2

Two aspects of new developments

  • Better QCD jet.
  • Smarter jet algorithm.
  • Noise suppression with jet grooming.
  • Jet substructure.
  • Boosted top.
  • Higgs.

Boston Jet Workshop: http://jets.physics.harvard.edu/workshop/Main.html Northwest Terascale workshop http://www.physics.uoregon.edu/~soper/Jets2011/talks.html Boost 2011, May, 23-27, Princeton. http://boost2011.org Boost 2012, July, 23-27, Valencia, Spain. http://ific.uv.es/boost2012/

2

Tuesday, September 11, 12

slide-3
SLIDE 3

Want to play with it?

  • Parton level Signal and background:

Madgraph, Alpgen, ...

  • ME+PS matching, UE, Pileup:

Pythia, Herwig, Sherpa, ...

  • Some detector effect, in particular, granularity

0.1x0.1

PGS, Delphes, “by hand”.

  • Jet tools.

Fastjet. SpartyJet

http://www.lpthe.jussieu.fr/~salam/fastjet/ http://projects.hepforge.org/spartyjet/

3

Tuesday, September 11, 12

slide-4
SLIDE 4

The importance of jets:

  • “Everywhere” at hadron colliders.
  • Present in (almost) all new physics signals.

Many of them only have hadronic channels.

˜ q ˜ q∗ ¯ q q ... ...

˜ q ˜ q∗ ¯ q q .... .... ¯ q q ˜ g ˜ g

jet jet jet jet jet jet

4

Tuesday, September 11, 12

slide-5
SLIDE 5

Jet look likes

  • When produced at TeV-scale energies, they have

a large boost.

Jets with substructure.

Challenge: distinguishing them from QCD jets (q and g). boost

q q top q ... q... W ± top b b

q, ... q, ... W ± q ... q... W ± ... q, ... ¯ q, ... Z q, ... ¯ q, ... Z

5

Tuesday, September 11, 12

slide-6
SLIDE 6

Need new jet tools for the LHC.

  • More energetic, bigger, jet at the LHC.

LHC jet: 50 GeV - several TeV Tevatron jet: 50 - 100s GeV

  • Much higher “noise” level at the LHC.
  • LHC: 10-100 GeV / rapidity
  • Tevatron: 2-10 GeV / rapidity

6

Tuesday, September 11, 12

slide-7
SLIDE 7

Hadron collision

Sketch of an event

Drawing: F. Krauss

7

Tuesday, September 11, 12

slide-8
SLIDE 8

Why can we calculate at all?

  • Perturbatively, we can only calculate with quark

and gluon in hard collisions.

  • Factorization.
  • IRC safety, need proper choice of observable.

Soft or collinear radiation should not be able to induce “large” changes in the observable. Otherwise, we cannot compare calculation with

  • bservables.

8

Tuesday, September 11, 12

slide-9
SLIDE 9

Factorization: intuition

9

proton

Tuesday, September 11, 12

slide-10
SLIDE 10

Factorization: intuition

9

Hard interaction time (distance) scale Q-1

proton

Tuesday, September 11, 12

slide-11
SLIDE 11

Factorization: intuition

9

Hard interaction time (distance) scale Q-1 “talking” to the rest of the proton time(distance) scale mproton-1

proton

Tuesday, September 11, 12

slide-12
SLIDE 12

Factorization: intuition

9

Hard interaction time (distance) scale Q-1 “talking” to the rest of the proton time(distance) scale mproton-1

If Q-1 ≪ mproton-1 (hard interaction) Two processes should not affect each other → Factorization!

proton

Tuesday, September 11, 12

slide-13
SLIDE 13

Factorization: intuition

9

Hard interaction time (distance) scale Q-1 “talking” to the rest of the proton time(distance) scale mproton-1

If Q-1 ≪ mproton-1 (hard interaction) Two processes should not affect each other → Factorization!

proton

A similar story for final states fragmentation, q,g⇒ hadrons (pion, K...)

Tuesday, September 11, 12

slide-14
SLIDE 14

Factorization

  • Schematics of production at hadron colliders.

2

?

Parton densities Threshold matrix elements

×phase-space

Partonic cross section

10

Tuesday, September 11, 12

slide-15
SLIDE 15

Hadron collision.

Sketch of an event

11

Hard interaction, gg⇒ g h t tbar ⇒h t tbar decay PDF PDF

Tuesday, September 11, 12

slide-16
SLIDE 16

Hadron collision

Sketch of an event

12

PDF PDF Clusters of hadronic energy final state object: jet pjet = Σ p of constituents Inclusive: independent of final states, just energy

Tuesday, September 11, 12

slide-17
SLIDE 17

Hadron collision

Sketch of an event

12

PDF PDF Clusters of hadronic energy final state object: jet pjet = Σ p of constituents Inclusive: independent of final states, just energy Very important: need pjet ≈ pparton Can use parton level calculation to predict jet properties

Tuesday, September 11, 12

slide-18
SLIDE 18

Hadron collision

Sketch of an event

13

PDF PDF Initial state radiation soft, long distance interactions Fragmentation (q,g ⇒ hadrons) ...

Tuesday, September 11, 12

slide-19
SLIDE 19

Factorization

S H

B1 B2 J1 J2

σ = B1 ⊗ B2 ⊗ H ⊗ J1 ⊗ J2 ⊗ S B1 = dx1 f(x1), B2 = dx2 f(x2). = H ⊗ J1 ⊗ J2 ⊗ S

14

Tuesday, September 11, 12

slide-20
SLIDE 20

Well tested.

2 3 4 5 6 [pb] ! 10

2

10

3

10

4

10

5

10

6

10 2 3 4 5 6 [pb] ! 10

2

10

3

10

4

10

5

10

6

10 ATLAS Preliminary

  • 1

L dt=2.43 pb

"

R=0.4, =7 TeV)+syst. s Data ( 1.11 × ALPGEN+HERWIG AUET1 0.65 × PYTHIA AMBT1 1.22 × ALPGEN+PYTHIA MC09’

Inclusive Jet Multiplicity 2 3 4 5 6 MC/Data 0.5 1 1.5 Inclusive Jet Multiplicity 2 3 4 5 6 MC/Data 0.5 1 1.5

100 200 300 400 500 600 700 800

[pb/GeV]

T

/d p ! d

  • 1

10 1 10

2

10

3

10

4

10

5

10

100 200 300 400 500 600 700 800

[pb/GeV]

T

/d p ! d

  • 1

10 1 10

2

10

3

10

4

10

5

10 ATLAS Preliminary

  • 1

L dt=2.43 pb

"

R=0.4, =7 TeV)+syst. s Data ( 1.11 × ALPGEN+HERWIG AUET1 0.65 × PYTHIA AMBT1 1.22 × ALPGEN+PYTHIA MC09’ 2 #

jets

N

(leading jet) [GeV]

T

p 100 200 300 400 500 600 700 800 MC/Data 0.5 1 1.5 (leading jet) [GeV]

T

p 100 200 300 400 500 600 700 800 MC/Data 0.5 1 1.5

ATLAS-CONF-2011-043

ATLAS-CONF-2011-043, 7 TeV, 2.43 pb-1

15

Tuesday, September 11, 12

slide-21
SLIDE 21

Why is it hard?

Tuesday, September 11, 12

slide-22
SLIDE 22

Why is it hard?

Tuesday, September 11, 12

slide-23
SLIDE 23

Why is it hard?

Tuesday, September 11, 12

slide-24
SLIDE 24

Why is it hard?

jet jet jet jet

Tuesday, September 11, 12

slide-25
SLIDE 25

Why is it hard?

Multiple interaction, underlying events, pile-up

jet jet jet jet “beam”

Tuesday, September 11, 12

slide-26
SLIDE 26

Why is it hard?

  • a. Overlapping jets.

Multiple interaction, underlying events, pile-up

jet jet jet jet “beam”

Tuesday, September 11, 12

slide-27
SLIDE 27

Why is it hard?

  • a. Overlapping jets.

ISR (beam) clustered

Multiple interaction, underlying events, pile-up

jet jet jet jet “beam”

Tuesday, September 11, 12

slide-28
SLIDE 28

Why is it hard?

  • a. Overlapping jets.

Part of the beam? ISR (beam) clustered

Multiple interaction, underlying events, pile-up

jet jet jet jet “beam”

Tuesday, September 11, 12

slide-29
SLIDE 29

Why is it hard?

  • a. Overlapping jets.

Proper choice of cone size? Part of the beam? ISR (beam) clustered

Multiple interaction, underlying events, pile-up

jet jet jet jet “beam”

Tuesday, September 11, 12

slide-30
SLIDE 30

Why is it hard?

  • To best preserve we would like to:
  • Use “smart” jet shapes.
  • Reduce “noise”.
  • a. Overlapping jets.

Proper choice of cone size? Part of the beam? ISR (beam) clustered

Multiple interaction, underlying events, pile-up

jet jet jet jet “beam”

Tuesday, September 11, 12

slide-31
SLIDE 31

What do jets look like?

17

Tuesday, September 11, 12

slide-32
SLIDE 32

Parton splitting, collinear limit

θ pM pA pB

z = EA EM t = p2

M = (pA + pB)2

The main feature of radiation can be seen by considering the Collinear limit: 휃 ⇒ 0, t≪ EM2

Relevant kinematical variables

φ

Tuesday, September 11, 12

slide-33
SLIDE 33

Collinear factorization

1 2 n − 2 n − 1 M 1 2 n − 2 n − 1 M A B A B A B |Mn+1|2 = |M(p1, ...pA, pB)|2 |Mn(p1, ...pM)|2 |M(pM → pApB)|2 ×

collinear limit

|Mn+1|2dΠn+1 ' |Mn|2dΠn dt t αS 2π P(z)dzdφ

Tuesday, September 11, 12

slide-34
SLIDE 34

Collinear factorization

1 2 n − 2 n − 1 M 1 2 n − 2 n − 1 M A B A B A B |Mn+1|2 = |M(p1, ...pA, pB)|2 |Mn(p1, ...pM)|2 |M(pM → pApB)|2 ×

collinear limit

|Mn+1|2dΠn+1 ' |Mn|2dΠn dt t αS 2π P(z)dzdφ

collinear singularity: t⇒0

Tuesday, September 11, 12

slide-35
SLIDE 35

Collinear factorization

1 2 n − 2 n − 1 M 1 2 n − 2 n − 1 M A B A B A B |Mn+1|2 = |M(p1, ...pA, pB)|2 |Mn(p1, ...pM)|2 |M(pM → pApB)|2 ×

collinear limit

|Mn+1|2dΠn+1 ' |Mn|2dΠn dt t αS 2π P(z)dzdφ

collinear singularity: t⇒0 Splitting function IR singularity: z⇒0, 1

P(z) ∝ |M(pM → pApB)|2

Tuesday, September 11, 12

slide-36
SLIDE 36

Splitting function, IR singular as z⇒0, 1

Pq→qg(z) = CF 1 + z2 1 − z , Pg→gg(z) = CA 1 − z z + z 1 − z + z(1 − z)

  • ,

Pg→q¯

q(z) = TR

  • z2 + (1 − z)2

,

Combining with

|Mn+1|2dΠn+1 ' |Mn|2dΠn dt t αS 2π P(z)dzdφ

Radiation wants to be collinear and soft

Tuesday, September 11, 12

slide-37
SLIDE 37

Shape of a jet: parton shower

  • From the initial parton, a jet is built up by many

radiations (splittings).

Prefers collinear radiation

P ~ (z)-1 prefers soft radiation

QCD jet: a cluster of radiation a) relatively soft b) close to the direction of PM c) approximately symmetrical around PM

21

QM2 = t

Tuesday, September 11, 12

slide-38
SLIDE 38

Jet Algorithms.

  • Two type of decisions, based on two types criteria:
  • What to cluster.
  • When to stop cluster.
  • Choice of the criteria determines the properties of the

jets.

A set of vectors {pi} Calorimeter towers... Jets: {PJ} Clustering

  • 6
  • 4
  • 2

2 4 6 1 2 3 4 5 6 0 5 10 15 20 25 SISCone, R=1, f=0.75 y [GeV]

t

p φ

22

Tuesday, September 11, 12

slide-39
SLIDE 39

First consideration: IRC safety

23

soft radiation collinear splitting

Tuesday, September 11, 12

slide-40
SLIDE 40

First consideration: IRC safety

Soft or collinear radiation should not be able to induce “large” changes in the observable. Otherwise we cannot compute and compare with experiments.

23

soft radiation collinear splitting

Tuesday, September 11, 12

slide-41
SLIDE 41

First consideration: IRC safety

Soft or collinear radiation should not be able to induce “large” changes in the observable. Otherwise we cannot compute and compare with experiments.

23

soft radiation collinear splitting

All of these should be 2 jet events!

Tuesday, September 11, 12

slide-42
SLIDE 42

Seeded cone

  • Starting with a set of seeds (momenta which are “more

likely” to be the centers of the jets).

  • Draw a cone of certain size around each seed.
  • Within each cone, add up all momenta. Use the new

direct as the new seed.

  • Iterate this process until we end up with stable cones.

24

Tuesday, September 11, 12

slide-43
SLIDE 43

Seeded Cone, IR unsafe

100 200 300 400 500 −1 1

pT (GeV/c)

stable cones from seeds 100 200 300 400 500 −1 1

pT (GeV/c)

add soft particle

100 200 300 400 500 −1 1

pT (GeV/c)

resolve overlaps

25 No other radiation with the radius of cones centered on the seeds “stable cone”, clustering stops. 2 jets.

an event with 2 jets becomes an event with one jet because of a soft radiation

Tuesday, September 11, 12

slide-44
SLIDE 44

Sequential recombination jet algorithm

  • Basic ingredients of a “sequential” jet algorithm.
  • Two types of “distances”
  • Jet-jet distance: “when to cluster”
  • Jet-beam distance: “when to stop clustering”
  • Pair wise comparison of all distances
  • If smallest distance at any stage in clustering is jet-jet,

add together corresponding four-momenta, else take jet with smallest jet-beam distance and set it aside.

  • Repeat till all jets are set aside.

dij

diB

26

Tuesday, September 11, 12

slide-45
SLIDE 45

d12 d23 d13

φ η

d12 < d13 < d23 < d(1,2,3)B < di4 4

27

Tuesday, September 11, 12

slide-46
SLIDE 46

φ η

4

28

Tuesday, September 11, 12

slide-47
SLIDE 47

d12

φ η

d12 < d(1,2)B < di4 4

29

Tuesday, September 11, 12

slide-48
SLIDE 48

φ η

4

30

Tuesday, September 11, 12

slide-49
SLIDE 49

φ η

4

Done!

31

Tuesday, September 11, 12

slide-50
SLIDE 50

Coordinate System

η = − ln ⇤ cot θ 2 ⇥⌅

32

Distance measure:

Tuesday, September 11, 12

slide-51
SLIDE 51

Recombination Algorithms

  • kT algorithm
  • C/A algorithm
  • anti-kT algorithm

dij = min(p2

T i, p2 T j)

∆R R0 ⇥2 , diB = p2

T i

dij = min(p−2

T i , p−2 T j )

∆R R0 ⇥2 , diB = p−2

T i

dij = ∆R R0 ⇥2 , diB = 1

A B

anti−kT

B A

C/A

B A

kT

33

Tuesday, September 11, 12

slide-52
SLIDE 52
  • 6
  • 4
  • 2

2 4 6 1 2 3 4 5 6 0 5 10 15 20 25

, R=1

t

k

y [GeV]

t

p φ

  • 6
  • 4
  • 2

2 4 6 1 2 3 4 5 6 0 5 10 15 20 25

Cam/Aachen, R=1

y [GeV]

t

p φ

  • 6
  • 4
  • 2

2 4 6 1 2 3 4 5 6 0 5 10 15 20 25 SISCone, R=1, f=0.75 y [GeV]

t

p φ

  • 6
  • 4
  • 2

2 4 6 1 2 3 4 5 6 0 5 10 15 20 25

, R=1

t

anti-k

y [GeV]

t

p φ

34

Tuesday, September 11, 12

slide-53
SLIDE 53

! Example: here’s an event with 500 GeV dijets (left), and

the same event with fifty pileup events (right).

! We’ll encounter this level of pileup next year, ! Somehow we’re going to have to find new physics in

this mess!

35

Messy environment

Some “clean up” procedure, filtering, pruning, trimming.

Tuesday, September 11, 12

slide-54
SLIDE 54

Shape of jets.

!"pt#2

pert + !"pt#2 h + !"pt#2 UE [GeV2]

R Tevatron quark jets pt = 50 GeV 1 2 3 4 5 6 7 8 9 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 !"pt#2

pert

!"pt#2

h

!"pt#2

UE

0.4 0.5 0.6 0.7 0.8 0.9 1 50 500 100 1000 best R pt [GeV] Tevatron, gluon jets Tevatron, quark jets LHC, gluon jets LHC, quark jets

“best” R

  • G. Salam, 0906.1833

Tuesday, September 11, 12

slide-55
SLIDE 55

Shape of jets.

!"pt#2

pert + !"pt#2 h + !"pt#2 UE [GeV2]

R Tevatron quark jets pt = 50 GeV 1 2 3 4 5 6 7 8 9 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 !"pt#2

pert

!"pt#2

h

!"pt#2

UE

0.4 0.5 0.6 0.7 0.8 0.9 1 50 500 100 1000 best R pt [GeV] Tevatron, gluon jets Tevatron, quark jets LHC, gluon jets LHC, quark jets

“best” R

  • G. Salam, 0906.1833

radiation out of the cone

Tuesday, September 11, 12

slide-56
SLIDE 56

Shape of jets.

!"pt#2

pert + !"pt#2 h + !"pt#2 UE [GeV2]

R Tevatron quark jets pt = 50 GeV 1 2 3 4 5 6 7 8 9 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 !"pt#2

pert

!"pt#2

h

!"pt#2

UE

0.4 0.5 0.6 0.7 0.8 0.9 1 50 500 100 1000 best R pt [GeV] Tevatron, gluon jets Tevatron, quark jets LHC, gluon jets LHC, quark jets

“best” R

  • G. Salam, 0906.1833

radiation out of the cone Universal noise ∝ R2

Tuesday, September 11, 12

slide-57
SLIDE 57

Going beyond anti-KT: “noise” control

  • Noise: Initial state radiation (ISR), multiple interaction

(MI), underlying events (UE), pile-up (PU).

Mass [GeV]

900 920 940 960 980 1000 1020 1040 1060 1080 1100

Cross Section [A.U.]

0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4

R=0.9 R=1.1 R=1.3 R=1.5

Mass [GeV]

900 920 940 960 980 1000 1020 1040 1060 1080 1100

Cross Section [A.U.]

0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16

R=0.9 R=1.1 R=1.3 R=1.5

FSR only Including ISR, MI, UE, pile-up

Room for improvement!

Tuesday, September 11, 12

slide-58
SLIDE 58

Jet trimming.

  • Introducing a “cut” on soft radiation.
  • Discard “stuff” below the cut after jet clustering.
  • Our implementation.
  • Cluster all calorimeter data using any algorithm
  • Take the constituents of each jet and recluster with

smaller radius Rsub (Rsub = 0.2 seems to work well).

  • Discard the subjet i if
  • Best choice of the hard scattering scale and fcut.
  • Process dependent.
  • Can be optimized experimentally.

pT i < fcut · Λhard

  • D. Krohn, J. Thaler, LTW, arXiv:0912.1342

ISR argument.

38

Tuesday, September 11, 12

slide-59
SLIDE 59

Why is it possible to gain?

  • MI, UE, and pile-up are incoherent soft background. They

can be effectively removed with a cut on soft radiation.

  • Both FSR (want to keep) and ISR (want to discard) have

soft radiation, but

  • ISR:
  • FSR is controlled by both collinear and soft

singularities:

  • Tends to be clustered into subjet, and kept.
  • Therefore, a soft cut relative to the jet energy flow

could enhance FSR relative to ISR.

39

Tuesday, September 11, 12

slide-60
SLIDE 60

! "

  • 1.5
  • 1
  • 0.5

0.5 1 1.5 # "

  • 1.5
  • 1
  • 0.5

0.5 1 1.5 ! "

  • 1.5
  • 1
  • 0.5

0.5 1 1.5 # "

  • 1.5
  • 1
  • 0.5

0.5 1 1.5 ! "

  • 1.5
  • 1
  • 0.5

0.5 1 1.5 # "

  • 1.5
  • 1
  • 0.5

0.5 1 1.5 ! "

  • 1.5
  • 1
  • 0.5

0.5 1 1.5 # "

  • 1.5
  • 1
  • 0.5

0.5 1 1.5

Start Cluster into subjets Discard soft subjets Reassemble 1 2 3 4

40

Tuesday, September 11, 12

slide-61
SLIDE 61

Simple test case: di-jet resonance

Improvement fcut, Ncut Rsub R0, ρ Γ [GeV] M [GeV] anti-kT

  • 1.0∗

71 522 anti-kT (N) 40% 5∗ 0.2∗ 1.5∗ 62 499 anti-kT (f, pT ) 59% 3 × 10−2∗ 0.2 1.5 52 475 anti-kT (f, H) 61% 1 × 10−2∗ 0.2 1.5 50 478 VR 30%

  • 200∗ GeV

62 511 VR (N) 53% 5 0.2 275∗ GeV 53 498 VR (f, pT ) 68% 3 × 10−2 0.2 300∗ GeV 49 475 VR (f, H) 73% 1 × 10−2 0.2 300∗ GeV 47 478 Filtering 27% 2 R0/2 1.3∗ 61 515

Mass [GeV]

400 420 440 460 480 500 520 540 560 580 600

Cross Section [A.U.]

0.05 0.1 0.15 0.2 0.25

T

anti-k trimmed

T

anti-k

Mass [GeV]

400 420 440 460 480 500 520 540 560 580 600

Cross Section [A.U.]

0.05 0.1 0.15 0.2 0.25

VR VR trimmed

  • We provide plugins fully compatible with Fastjet.

http://jthaler.net/jets/VR_Jets.html http://jthaler.net/jets/Jet_Trimming.html

41

Tuesday, September 11, 12

slide-62
SLIDE 62

Jet mass: help from new jet algorithm

  • Effect of radiation contamination on the jet mass
  • Trimming gives large improvement by reducing effective

jet size significantly.

More faithful (smaller) jet mass for the background.

Jet Mass [GeV]

20 40 60 80 100 120 140 160 180 200

Cross Section [A.U.]

100 200 300 400 500

FSR only

T

anti-k ISR/MI/pileup

T

anti-k Trimming FSR only Trimming ISR/MI/pileup

Without contamination With “trimming” With contamination

Tuesday, September 11, 12

slide-63
SLIDE 63

Jet substructure

43

Tuesday, September 11, 12

slide-64
SLIDE 64

X ¯ t t

When to consider substructure

  • Have to consider the boosted objects.
  • It is beneficial to consider the boosted objects.

e.g. Lower combinatorics, SM background boost differently. W, Z b b

For example, boost tops Brooijmans; Lillie, Randall, LTW; Thaler, LTW;

  • D. Kaplan, K. Reherman, M. Schwartz, B. Tweedie;
  • L. Almeida, S. Lee, G. Perez, G. Sterman, I. Sung, J.

Virzi

  • S. Chekanov and J. Proudfoot.

... Butterworth, Davidson, Bubin, Salam

For a summary of recent developments: C. Vermilion,1001.1335 e.g. h

Tuesday, September 11, 12

slide-65
SLIDE 65

Shape of a jet: parton shower

  • From the initial parton, a jet is built up by many

radiations.

Prefers collinear radiation

P ~ (z)-1 prefers soft radiation

QCD jet: a cluster of radiation a) relatively soft b) close to the direction of PM c) approximately symmetrical around PM

45

Tuesday, September 11, 12

slide-66
SLIDE 66

(hadronic) Top tagging at the LHC

  • Fully collimated tops look like QCD jets.

W + b t q E1 E2

Zooming in near the first splitting Soft radiation: Top. First splitting Jet mass: QCD. Decay: Jet mass:

  • QCD: radiation.
  • Top decay: 3 hard objects.

Basic distinction:

46

Tuesday, September 11, 12

slide-67
SLIDE 67

Building a microscope to look inside jets.

  • The jet clustering history is approximately the inverse of

the parton shower.

!1'2-/.-)*#3*+,-*4-+*5$6.+-/&'(*,&.+#/07

47

Tuesday, September 11, 12

slide-68
SLIDE 68

Top jets vs QCD jets

  • Combined cuts on jet mass and z can enhance further

the signal with respect to the background.

∼

48

Tuesday, September 11, 12

slide-69
SLIDE 69

More jet shape variables.

  • Top decay is more like 3-body. Span a “plane”

perpendicular to the jet axis.

  • Transverse sphericity, or planar flow

Thaler and LTW, arXiv:0806.0023. Almeida, Lee, Perez, Sterman, Sung, Virzi, arXiv:0807.0234

Pf ≈ 0 Pf ≈ 1

Ikl

w =

⇤

i

wi pi,k wi pi,l wi

Pf = 4λ1λ2 (λ1 + λ2)2

49

Tuesday, September 11, 12

slide-70
SLIDE 70

Grooming gives better jet shape

  • Can be used to further improve top tagging. An

additional factor of several possible.

  • Interesting to compare with improved QCD calculation,

using modern technologies such as SCET.

Planar Flow

0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Cross Section [A.U.]

100 200 300 400 500

FSR only

T

anti-k ISR/MI/pileup

T

anti-k Trimming FSR only Trimming ISR/MI/pileup

Planar flow

With “trimming” With no contamination With contamination

50

Tuesday, September 11, 12

slide-71
SLIDE 71

Various top taggers

  • G. Brooijmans, arXiv:0802.3715;

CMS Coll. CMS PAS JME-09-001

  • J. Thaler, LTW, arXiv:0806.0023
  • D. Kaplan, K. Reherman, M. Schwartz, B. Tweedie, arXiv: 0806.0848.
  • L. Almeida, S. Lee, G. Perez, G. Sterman, I. Sung, J.

Virzi, arXiv:0807.0243

  • L. Almeida, S. Lee, G. Perez, G. Sterman, I. Sung, arXiv:1006.2035

Barger, Huang, 1102.3183

efficiency 0.1 0.2 0.3 0.4 0.5 0.6 0.7 mistag rate

  • 3

10

  • 2

10

  • 1

10 1

Hopkins CMS Pruning ATLAS Thaler/Wang

Boost 2010 proceeding, 1012.5421

51

Tuesday, September 11, 12

slide-72
SLIDE 72

BDRS (+filtering)

  • Z+H and W+H with H->bb.
  • Considered boosted Higgs.
  • Better acceptance.
  • backgroud such as ttbar

boost differently.

  • Similar result reproduced by

ATLAS

Mass (GeV)

20 40 60 80 100 120 140 160 180 200

  • 1

Events / 8GeV / 30fb

20 40 60 80 100 120 140

Mass (GeV)

20 40 60 80 100 120 140 160 180 200

  • 1

Events / 8GeV / 30fb

20 40 60 80 100 120 140

q q V+jets VV V+Higgs

= 4.5 B S/ in 112-128GeV

(d)

Butterworth, Davison, Rubin, Salam, 0802.2470

]

2

Higgs mass [GeV/c 20 40 60 80 100 120 140 160 180 200

  • 1

Events / 8GeV / 30fb 10 20 30 40 50 ]

2

Higgs mass [GeV/c 20 40 60 80 100 120 140 160 180 200

  • 1

Events / 8GeV / 30fb 10 20 30 40 50

tt V+jets VV Higgs

Total S = 16.3 B = 104.2 Range 104-136GeV

ATLAS preliminary

(simulation) (c)

ATL-PHYS-PUB-2009-088

52

Tuesday, September 11, 12

slide-73
SLIDE 73

New developments: N-jettiness

  • Using event shape instead of clustered jets.
  • Allowing better QCD (SCET) treatments.
  • Example: application in jet veto in Higgs searches.

TN = ⇤

k

| pkT | min

  • da(pk), db(pk), d1(pk), d2(pk), . . . , dN(pk)

⇥ ≡ T a

N + T b N + T 1 N + · · · + T N N Jet 2 Jet b Jet a Soft Jet 3 Jet 1

b a 1 3 2

p p − +

(c) pp → leptons plus jets.

Stewart, Tackmann, Waalewijn, 1004.2489 N-jet like event

Tuesday, September 11, 12

slide-74
SLIDE 74

Superstructure

  • Using more global information.
  • Applications to other channels as well.

pp->H->bb pp->g->bb

Gallicchio, Schwartz, 1001.5027 Relative enhanced radiation consentration e.g., ttbar at Dzero, Haas Boston Jet Workshop

Tuesday, September 11, 12

slide-75
SLIDE 75

Unbury the Higgs.

F

  • r

e x a m p l e : B . B e l l a z z i n i , C . C s a k i , A . F a l k

  • w

s k i , A . W e i l e r , a r X i v : 9 1 . 3 2 1 , a r X i v : 9 6 . 3 2 6 For example: P . Graham, A. Pierce, J. Wacker, hep-ph/0605162

  • M. Carena, T. Han, G. Huang, C. Wagner, arXiv:0712.2466

h

a a

Soft gluon jets, considered impossible.

55

Tuesday, September 11, 12

slide-76
SLIDE 76

Unbury the Higgs.

F

  • r

e x a m p l e : B . B e l l a z z i n i , C . C s a k i , A . F a l k

  • w

s k i , A . W e i l e r , a r X i v : 9 1 . 3 2 1 , a r X i v : 9 6 . 3 2 6 For example: P . Graham, A. Pierce, J. Wacker, hep-ph/0605162

  • M. Carena, T. Han, G. Huang, C. Wagner, arXiv:0712.2466

Two “equal ”clusters

h

Less radiation

  • utside this cone

Boosting the Higgs.

56

Tuesday, September 11, 12

slide-77
SLIDE 77

Encouraging results.

0.5 1 1.5 2 2.5 60 70 80 90 100 110 120 130 140

Cross Section [fb/10-GeV] Mass [GeV]

Signal Background 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 60 70 80 90 100 110 120 130 140

Cross Section [fb/10-GeV] Mass [GeV]

Signal Background

W/Z+h Chen, Nojiri, Sreethawong, 1006.1151 Falkowski, Krohn, Shelton, Thalapillil, and LTW, 1006.1650 ttbar+h

57

Tuesday, September 11, 12

slide-78
SLIDE 78

Substructure can also be useful for

  • From top/W/Z/Higgs from NP decay, early LHC prospects.
  • Resonance ttbar.
  • SUSY.
  • Top partner to Higgs.
  • Z’ to WW, Zh...
  • Boosted NP particles.
  • Neutralino + RPV
  • Boosted gluino from squark.

Kribs, Martin, Roy, Spannowsky , 0912.4731, 1006.1656 Kribs, Martin, and Roy, 1012.2886 Cui, Han, Schwartz, 1012.2077 Katz, Son, Tweedie, 1010.5253 Butterworth, Ellis, Raklev, Salam, 0906.0728 Fan, Krohn, Mosteiro, Thalapillil, 1102.0302

Tuesday, September 11, 12