1 / 62

Introduction to Jet Substructure

大连大学. Introduction to Jet Substructure. Shuo Yang (杨硕) Physics Department, Dalian University/ CHEP , Peking University. First two years at the LHC 2012.06.13. Outline. Large Hadron collider What is jet. Jet cone size and fat jet. Jet substructure

maille
Download Presentation

Introduction to Jet Substructure

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. 大连大学 Introduction to Jet Substructure Shuo Yang (杨硕) Physics Department, Dalian University/ CHEP,Peking University First two years at the LHC 2012.06.13

  2. Outline • Large Hadron collider • What is jet. • Jet cone size and fat jet. • Jet substructure Top taggers and Higgs taggers Grooming Hybrid-R—heavy charged Higgs • Summary

  3. LHC

  4. Large Hadron Collider • Incoming proton • Parton distribution function 2. Hard process Matrix element calculation , LO or NLO 3. Radiation (Parton Shower) Initial state radiation (ISR) and Final state radiation (FSR) p p 4. Hadronization Universal models 5. Underlying event

  5. I. What is jet? quark quark gluon quark quark hadronisation Why do we see jets? Parton fragmentation

  6. Jet Definition • Which? • How? Jet algorithms provide a set of rules for grouping particles into jets. + Parameters Recombination schemes define how to combine their momenta . (mostly E-scheme)

  7. Jet Algorithms Cone Algorithm Recombination Algorithm UAI,JetCLU,SISCone…. Jade,Kt, Cambrige-Aachen, Anti-Kt…. Based on fixed geometry Idea: Put cone along dominatant direction of flow direction Idea: Undo branching

  8. Calculate the distance dij between all pairs of pseudo-jetsand the beam distance diB for each pseudo-jet. • Find the minimum in the list of dij and diB . • If the minimum is dij, recombine i and j into a new particle and then return to step 1. • If it is diB, define it as a final-state jet and remove it from the jet list. Return to step 1. • Iterate this process until the original list is empty, i.e., all peudo-jets have been clustered to jets. The dijanddiBin most popular sequential recombine algorithms for use at the LHC can be parameterized as: P is a parameter, 1 for kt, -1 for anti-kt, 0 for CA

  9. Cartoon for the Kt algorithm

  10. Cartoon for the Kt algorithm

  11. Cartoon for the Kt algorithm

  12. Cartoon for the Kt algorithm

  13. Cartoon for the Kt algorithm

  14. Cartoon for the Kt algorithm

  15. Cartoon for the Kt algorithm

  16. Cartoon for the Kt algorithm

  17. Cartoon for the Kt algorithm A 2 jets event

  18. Borrowed from Salam’s talk

  19. Remarks • There is no standard jet algorithms. The jet algorithm and the parameters could be chosen flexibly. • All these recombination algorithms are infrared-safe and collinear safe. • The anti-kt algorithms gets circular jets area which act much like an idealized cone-based algorithm, while the kt and CA get irregular jets area. • The kt and CA algorithms incline to combine the harder pseudo-jets or pseudo-jets with a large separation in the later, so it is expected that the combination in the last step is the reverse of the main QCD branching or the 2-body decay of a massive particle. So decluster can be used in CA and kt in substructure study.

  20. Boosted Massive Particles Lab frame Rest frame Hadronically decaying EW massive particles at high Pt = 2 jets R=0.7 R=0.4

  21. Jet cone size and fat jet R=1.5 R=1.0 R=0.5

  22. small cone vs. large cone perturbative fragmentation non-perturbative fragmentation underlying event & pile up multi-hard-parton

  23. Fat Jet • Cone size should be chosen flexibly for physical intention. • Boosted massive hadronically decaying objects is common at the LHC. M_Z/TeV • Fat jet is useful to capture hadronical decay of the highly boosted massive particle. W, H, t • Generally, mass discriminator is not enough! In order to separate signal, one have to anatomize the Jet Substructure and reconstruct delicate kinematical variables on subjets !

  24. 今天您吃了吗? 今天您收集事例了吗 吃了! 收集了! • 您吃的什么? 收集到什么事例了? 爪子 (蹄子)jets 什么爪子? 几个指头什么样的jets,几个subjets

  25. Brief history of Jet Substructure • Michael H. Seymour Searches for new particles using cone and cluster jet algorithms: A Comparative study, Jun 1993, Z. Phys. C62(1994) 127-138 • Butterworth, Cox, Forshaw, Jan 2002 WW scattering at the CERN LHC, Phys. Rev. D65 (2002) 096014 • Butterworth, Davison, Rubin, Salam, Feb 2008, Jet substructure as new Higgs search channel at the LHC, Phys.Rev.Lett.100(2008) 242001. • Kaplan, Rehermann, Schwartz and Tweedie, ``Top Tagging: A Method for Identifying Boosted Hadronically Decaying Top Quarks,'' Phys. Rev. Lett.101(2008) 142001 (“JH top-tagger”) • Filtering ,Trimming, Pruning , Variable R, NSubjettiness……. • Higgs –tagger, top-tagger, W-tagger ….. • H.N.Li, Z. Li, C.-P. Yuan, Phys.Rev.Lett.107(2011)152001 • Jet shape …… ………………

  26. Higgs taggers • Top taggers • Grooming • Searching for heavy charged Higgs. Hybrid-R

  27. ppVH Hbb Go to high transverse momentum !? Only 5% of the total events.(Pt>200GeV) Kill top pair background? Gain significance? Large background gg tt lvbbjj

  28. Borrowed from G.P. Salam’s talk

  29. Borrowed from G.P. Salam’s talk

  30. JH Top Tagger PRL101 (2008)142001

  31. Grooming Research by Bionsen, a natural deodorant company, found that the average woman's daily grooming and make-up routine means she 'hosts' a staggering 515 different synthetic chemicals on her body every single day.

  32. Dog grooming If grooming can be used for human and dogs, one is led to ask whether it can be used for jets.

  33. Jet Grooming • Filtering • Trimming • Pruning 1. Grooming narrows the mass distributions of jet and the reconstructed massive particles. 2. It improves both signal purity and the signal significance. 3. It can be used to refine any jet. 4. Grooming is time consuming but is promising.

  34. Filtering Figures are borrowed from G.P. Salam’s talk

  35. Refine the jet resolution • Filtering: break jet into subjets on angular scale Rfilt , take n filtered hardest subjets. • Trimming: break jet into subjets on angular scale Rtrim, take all subjets with Pt,sub>εPt,jet • Pruning: If the subjets about to be recombined have ΔR>Rprune and min(Pt1,Pt2)< ε(Pt1+Pt2), discard the softer one.

  36. Searching for heavy charged Higgs with jet substructure Shuo Yang, Qi-Shu Yan JHEP02(2012)074 arXiv: 1111.4530 taggered top

  37. Hybrid-R Reconstruction Method • CA jet algorithm with a large cone and top tagger to capture the highly boosted top t1. Form a jet list J0 . • Recluster the pseudo-jets in the event with a small cone (anti-kt R=0.4) and then get a new jet list L0 . • If a small-size jet is within the larger cone of the direction of the tagged top jet, remove it and then form a new jet list L1 . • We identify the most energetic jet in L1 as the b3 jet. The unused jets in the jet list form L2 . • We require there are at lest 3 jets in the list L2. Reconstruct the second top and W by +

  38. 1 2 3 4

  39. R size vs. mass distribution

More Related