Higgs Session Notes


April 11, 10.30 am, Jim Olsen: Review of SM Higgs Physics--Mass and Spin
Comments:
  • Mass:
    • clarification on bias study in mass measurement. Included a spurious signal and tested alternative background shapes
    • optimization of analysis not do to reduce overall uncertainty but with the understanding of systematic effects in mind
    • analyses by ATLAS and CMS were not modified for the combination. Uncertainties were grouped to allow comparisons of impact on the results.

  • Spin and parity
    • what scales of new physics are probed by β€œparity measurement"?
    • is the underlying Lagrangian for the amplitude fits well defined?
    • what precision is possible with VBF and Higgs to tau tau?
    • is there sensitivity to parton showering? 2nd order effect


April 11, 1:45 pm, Eilam Gross: Review of SM Higgs Physics--Couplings

Comments:
  • discussion on the correlation of the PDF uncertainty between VBF and VH. The assumption in the fit is 100% correlation while the uncertainties are ~60% correlated. The assumption is made for technical reasons.
  • Why is the VBF Higgs to bb CMS result not included in the combination? ATLAS does not have this measurement and the sensitivity is very small
  • Longer discussion on the sign of couplings. Interference terms give sensitivity to the sign. Most prominent in hgg, but also in ZH, single top, etc.
  • Conclusions: 1st and 2nd generation are not there;
  • please be careful with correlation matrices... Recast is a tricky business. We will try to issue a note and a code on how to do recast (EG)


April 12, 10:00 am, Markus Klute: Review of BSM Higgs Physics: Extended Higgs Sectors

Comments and questions:

  • An item to add to the menu--search for CP-violating observables mediated by Higgs exchange.
  • Question on the sensitivity of 750 GeV resonance in the Z𝛾 channel (both experimental and in theoretical models).
  • Poor background modeling in the ATLAS search for heavy charged Higgs (contrast with CMS where no excesses are seen in this channel).
  • No searches for H+ β†’ W+ 𝛾 (this is a one-loop process); may have been relegated to the Exotics group.
  • Two 2𝜎 excesses in ATLAS A β†’ Zh search (for A masses of roughly 300 GeV and 450 GeV). Does CMS see anything comparable?
  • The limits shown in the m(A) vs. tan 𝛽 plane based on searches for H and A of the MSSM employ a benchmark scenario in which both H and A can have significant branching ratios into neutralino/chargino pairs. As a result, the branching ratio into 𝜏+𝜏- is suppressed (compared to scenarios in which supersymmetric particle decays of H and A are absent). ATLAS and CMS should also provide limits in the m(A) vs. tan 𝛽 plane assuming that H and A decay only into SM particles (which would require a small tweak on the benchmark scenarios currently in use).
  • In the exact alignment limit of the 2HDM, the search for heavier Higgs states in the 𝛾𝛾 channel can yield significant constraints on the 2HDM parameter space.
  • The search for A/H β†’ tt in the large scalar mass, moderate tan 𝛽 region (the "LHC wedge") has been proposed as a way to search for heavy Higgs states in this parameter regime by Djouadi and collaborators. However, their analysis does not take into account important interference effects with continuum tt production, and thus the sensitivity of this search channel is less impressive than originally claimed. Are there any other ways to attack the LHC wedge region?
  • What is the status of the search for the charged Higgs boson via the cs final state. Consider both the case of charged Higgs masses below and above the mass of the top quark. Is there theoretical motivation that suggests that the cs final state may be relevant above the tb threshold? (One probably needs an usual flavor structure in the charged Higgs interactions).
  • If the charged Higgs mass is very close to the top mass, then current experimental Higgs searches are typically not applicable. Moreover, there does not seem to be a complete NLO analysis of charged Higgs production in this regime.
  • In searches for additional scalar states of an extended Higgs sector, many SM backgrounds are at play. It would be very useful to provide individual backgrounds for each search that could be used later in other contexts.
  • There was a general plea from theorists for more documentation on the many searches so that previous analyses could be revisited and reproduced at a later time. How do do this in practice led to a long discussion. It was suggested that to carry this out in practice (given limited budgets and time) might imply that fewer searches could be carried out.


April 12, 11:00 am, William Murray: Review of BSM Higgs Physics: Exotics

Comments and questions:
  • Concerning the Higgs portal. Can you really combine the constraints from dark matter direct detection and LHC searches? How much model dependence is involved? Part of the answer depends on cases where the relevant interactions can be expressed in terms of renormalizable or higher dimension operators.
  • In light of the bounds on Higgs to invisible decays, how much can be said about the possibility of h β†’ πœ’πœ’ in the MSSM where πœ’ is the LSP.
  • What type of (UV complete) models yield invisible final state particles with mass of about 1 GeV which could appear in the final state of invisible Higgs decays? Are there cases in which such low mass states exist that could show up in direct detection experiments in searches for dark matter? (That is the corresponding WIMP nucleon cross section is not too small.)
  • Is there strong theoretical motivation for the decay of the Higgs boson into long lived particles that could show up in LHC searches as displaced vertices?
  • In many of the LHC analyses, one allows for the possibility of the observed Higgs boson to decay into exotic final states (e.g. invisible particles or unusual final states made up of SM particles). The branching ratio into exotics is parameterized by one number. How can one best take into account the possibility of exotic production mechanisms for the observed Higgs boson?
  • It was pointed out that the measurement of H->WW is sensitive to the product of the HWW coupling and the width of the W boson. Normally this is fixed to SM
  • A huge issue associated with BSM Higgs physics is the need for accurate modeling of top quark production in association with jets and/or multi-leptons. There is a cumulative effect of the numerous uncertainties associated with these signals.
  • How good is the modeling of ttbb final states? Is there a comparison of various tools available? Te Higgs Cross Section Working Group has addressed this issue in the past, but perhaps it is time to revisit.
  • Another possible way to access the Higgs to invisible decays is via monojet events (e.g., gg β†’ hg, where he decays invisibly). How effective is such an analysis (how well can one separate from Z+jet, where Z decays into neutrinos?).


April 12, 2:00 pm, Pierre Savard: Review of Prospects for di-Higgs Production

Comments and questions

  • Can the 𝜏+𝜏-𝜏+𝜏- channel be used in searches for hh production?
  • In the excluded 2HDM regions in the tan 𝛽 vs, cos(𝛽-𝛼) plane, it would be helpful to overlay the constraints imposed by the "precision" measurements of the 125 GeV Higgs boson, interpreted as one of the neutral Higgs bosons of the 2HDM.
  • What precision should one aim to get for the Higgs self-coupling parameter πœ†? That is, what level of precision is needed in various theoretical scenarios? Among the scenarios to be considered is one where the observed Higgs boson at 125 GeV has SM-like Higgs self-couplings.
  • In the experimental search for the hhh vertex, consider the bbπœ‡+πœ‡- channel at the high-luminolsity LHC.
  • In order to extract a value for the Higgs self-coupling parameter πœ†, what is the best way to proceed? One can regard this as a consistency test of the SM (in which case you are trying to verify the consistency of the data with a measured value of πœ† given by the SM). One can also generalize the πœ… formalism to include a possible deviation of the Higgs self-coupling parameter from its SM value. Finally, one can treat πœ† as a parameter of some EFT that contains higher dimensional operators. In the latter case, what is the proper consistent treatment and what are the relevant experimental observables?
  • In hh searches where h β†’ ZZ, can one make use of the hadronic channels in a ZZZZ final state?
  • In hh searches, what is the most effective way to kill the top quark backgrounds?
  • How viable is the detection of hh at the LHC via the qq' β†’W β†’ WHH process?
  • How viable is the detection of hh at the LHC using the bbbb final state via the VBF production mechanism (WW β†’ h β†’ hh, etc.).
  • For values of πœ† that deviate significantly from the SM value, the kinematics of the final state decay productions in hh production may be significantly altered. Is this effect properly taken into account in the proposed LHC searches for hh?
  • If non-SM contributions to hh production exist, can this in turn effect significantly the predicted values for single h production? For example, if an effective gghh operator existed that would modify the predicted cross-section for hh production, one could also use this operator to generate an anomalous contribution to gg β†’ h via gg β†’ hh β†’ h. Thus, how much does the observed Higgs data (properties of h(125) already measured) constrain anomalous contributions to hh production?


April 13, 1:00 pm, Timothy Barklow (SLAC): Review of Higgs studies at future lepton colliders
  • Comments and questions
    • The design shows a 54 km booster and a colliding ring? Yes, this allows continuous top-up
    • The original idea was to build an ILC tunnel long enough for the full energy and then add cavities for the upgrade: this has changed? Yes, that was the original idea to go from 250 to 500. Now we want to go to straight to 500, given the plans with circular machines
    • why would you run the 500 machine at 250? To obtain optimal results for couplings measurements
    • The machine can reduce backgrounds with polarization of both beams (up to 30% for e+ and 80% foe e-. This allows us to improve on inv. width
    • Some projections on a 750 GeV particle were shown. How can you guess the xs at 750 GeV? An audience member suggested that anyway it is an imaginary particle at an imaginary collider πŸ˜ƒ Note from editor: i*i makes the result real.
    • ILC compared to CEPC -> should mention in such a comparison that the CEPC is a stepping stone, and part of a larger program.
    • Motivation to go for an absolute measurement? Things not charged under the SM can be nailed at e+e- that won’t show up in ratios
    • question regarding realism of going for a first generation measurement (electrons), is the energy spread realistic? The answer is that the estimates are work in progress


April 13, 2:00 pm, Heather Gray (CERN): Review of Higgs studies at future hadron colliders

  • Comments and questions
    • Given that these machine are planned for far in the future, shouldn’t we discuss accelerator techniques (e.g. plasma acceleration)? We need both energy and luminosity and it is hard to see how you get high lumi with these potential future techniques. In any case, the development of these will take years
    • There was a question why Htj gets a 252 factor increase. Typo?
    • Somebody made the point that it is important to measure how well we’d do vs pT, to search for contributions BSM. This is an advantage of the high energy machines
    • How does kappa_t at HL-LHC compared to 1-2% FCC. Estimate was that HL-LHC would be in the range of 5-10% (to be checked)
    • Would we understand tty well enough? Note that it is already measured in Run 1 and precision will improve a lot at the LHC already
    • There was a nice point made on how to get the Higgs width from tttt
    • Question on whether the NNLO heavy quark is ok at 100 TeV (hh). Answer: we think so...


April 14, 11:00 am, Frank Krauss: Precision Calculations and Simulations

  • Main theme of the discussion: going beyond black-box: need to understand uncertainties and systematics
  • Have to be careful with the tools: can deliver more but have to question the resultsl. ex: what is the meaning of varying the scale by a favor 2 if you pick the wrong scale at first place.
    • Regarding assignment of scales, Cacciari et al bayesian method is an alternative

  • Discussion on the multi-jet merging and pt_gamma gamma plot shown: is the agreement luck? good agreement seen in many other cases not just this example
  • Comparison with fixed order predictions: need to understand full kinematic aspects
  • half-time summary
    • users of code will need training for the new tools (danger of using a more complicated back box)
    • can we make the tools easier to use, quicker, robust
  • overtime points
    • Comment on whether new physics would make these obsolete (see overtime slide)? We would still need precision eventually but we need to be pragmatic and spend our time where it makes a difference and match the needs of the experiments.
    • comment by Carlos on whether this type of precision is needed given estimations that use sidebands: yes, since the extrapolation from the sidebands is what typically drives the systematics uncertainty on the background estimate
    • Lance: sometimes sidebands fail but information not propagated?


April 15, 11:00 am, Carlos Wagner: Discussions on weakly coupled BSM Higgs

Comments and questions

  • Two approaches to interpreting Higgs couplings: (1) still room for deviations from SM behavior. Take pattern of deviations seriously. (2) The observed Higgs boson is SM-like, implying the alignment limit. Consider possible BSM phenomena consistent with the alignment limit.
  • Current data consistent with an enhanced Higgs coupling to top quarks and a suppressed Higgs coupling to bottom quarks. However, there is no indication of an enhanced gluon fusion production cross section, even though an enhanced top quark Higgs boson coupling would naively lead to an enhanced gluon fusion cross section, which is governed by the top quark loop.
  • In BSM models, how can one enhance πœ…(t) while suppressing πœ…(b) and πœ…(g)? Answer: πœ…(g) is a loop-induced quantity as compared to πœ…(t) and πœ…(b), whose leading contributions in extended Higgs sector (due to Higgs mixing) arise at tree-level. Hence, find new loop contributions that can reduce the
    value of πœ…(g).
  • Example: a SUSY model with large stop mixing and light stops can reduce the value of πœ…(g) from its SM value. In practice, this is not possible in the MSSM but is possible in extended SUSY models (e.g. the NMSSM). Two reasons--need to get the right radiatively corrected light Higgs mass and need a large enough value of tan𝛽 (say of order 10) in an approximate alignment limit to sufficiently modify both the top and bottom couplings of the Higgs boson. But values of tan𝛽 of order 10 along with H and A masses in the vicinity of 200--300 GeV are ruled out by ATLAS and CMS searches for H,A β†’ 𝜏+𝜏-.
  • Note: near the alignment limit, it is possible to suppress the bottom couplings of the Higgs boson at lower values of tan𝛽, but not with the required enhancement of the top quark--Higgs boson couplings.
  • Question: can these light stops also significantly modify other Higgs couplings via radiative effects?
  • Question: does the picture change if the Higgs couplings to quarks have a non-trivial flavor structure?
  • CMS sees a signal for h β†’ 𝜏 πœ‡ at the 2𝜎 level. The corresponding ATLAS signal in the hadronic channel is only 1𝜎, with an update to the other channels
    expected any day now. Models to explain the h β†’ 𝜏 πœ‡ signal at the observed level are not so easy to come by.
  • A SUSY model of Aloni, Nir and Stamou to explain the h β†’ 𝜏 πœ‡ signal is intriguing, although Carlos insists that this is not viable within the MSSM.
  • Comments on hh production--interesting connections of the deviation of the hhh coupling to the possibility of a 1st order electroweak phase transition.
  • The alignment limit of the MSSM and NMSSM. Easier in the latter; still possible in the former. But H β†’ hh is suppressed in the approach to the alignment limit.
  • In the case of the NMSSM, additional singlet-like Higgs states (hs) add new important signatures. In particular, H β†’ h hs, H β†’ hs hs and A β†’ h_s Z provide new opportunities for LHC searches. So, in conducting these searches, go beyond the searches for H β†’ hh and A β†’ hZ, by relaxing the assumption that the mass of the final state Higgs boson is at 125 GeV.

April 15, 2:00 pm, Kaustubh Agashe: Discussions on strongly coupled BSM Higgs
Comments and questions

  • Composite Higgs models offer a continuous deformation from SM to technicolor model: xi=v^2/f^2 is zero for SM and 1 for TC
  • Generic predictions: Higgs couplings receive corrections of order xi
    • it is not the discovery of the Higgs boson that killed TC
    • but rather the measurements of the Higgs couplings in rather good agreement with SM predictions
  • Current constraints
    • from EW precision test: v^2/f^2<0.1, i.e. f>600-700 GeV
    • from current Higgs coupling measurements: v^2/f^2<0.2, i.e. f>500-600 GeV
    • from future Higgs measurements at HL-LHC: v^2/f^2<0.05, i.e. f>1.1 TeV
  • Other predictions:
    • partly composite gluons: color octet spin-1 resonances with a mass ~ 3 TeV
    • partly composite W,Z: W', Z' resonances with a mass ~ 3 TeV (-> no conflict with EW precision tests)
    • partly composite top: top partners. we do expect light top partners (mT'~2f)
      • EW precision tests were already giving a lower bound (mT'>1.2TeV) : not a surprise that we didn't find them at run-1
      • HL-LHC: will probe up to mT' > 2TeV
      • pair production (model independent, QCD coupling) vs single production (model dependent, mixing with top and bottom)
      • Question: ATLAS and CMS are searching below 1TeV, what do we learn?
        • 1. the heavier, the more-tuned the model.
        • 2. the estimate mT'~2f is only indicative -> good to perform direct searches
  • Questions:
    • Flavor constraints were already difficult for TC, what the status for composite Higgs?
      • in TC: quadratic mixing -> 4 fermion operators at the origin of top mass is irrelevant
      • in composite Higgs: partial compositeness, ie linear mixing: alleviates flavor tension
    • If we don't wee anything at 100TeV, does it kill composite models?
      • no: it only pushes the models into more fine-tuned regions
    • Have all strongly coupled models an holographic duals?
    • no, small N model don't a priori
    • The Lore in the experiments is that VLQ should have mass below 1 TeV
      • It was part of the point of this talk to make it clear this need not be so.

April 18, 11:00 am, Howard Haber: Revisiting the Higgs Whishlist
  • what will be our Higgs wishlist moving forward?
  • goal for Thursday to generate a new list (the previous list is given below)
  • temptation to drive a square peg into a round hole
  • how well can we do second generation Higgs couplings; about 3 sigma with 100/fb and two experiments

The Higgs Wishlist of December, 2012

After a two week long Higgs Identification Rapid Response Workshop in December, 2012 at the KITP, the participants created a Higgs wishlist for future studies. It may be of interest to revisit this list, eliminate the items that have since been resolved, and perhaps add new items. Thus, for further consideration, I present the December, 2012 wishlist (suitably edited).
  • Is the observable Higgs state at 125 GeV responsible for the unitarization of W_L W_L scattering?
  • How close to the alignment limit is the 125 GeV Higgs boson?
    • Are there new Higgs sector phenomena in reach by experiment (alignment without decoupling?)
    • Can we distinguish between tree-level Higgs mixing and loop-induced effects due to BSM physics?
  • What if deviations from SM Higgs couplings are found?
    • If large deviations are detected, is there a compelling source of BSM physics that can account for the deviations?
    • If small deviations are eventually established, what are the systematics of the deviations, and do they point to a particular BSM scenario and/or extended Higgs sector?
  • Concerning the use of precision Higgs observables as a probe of new physics:
    • How well can the High-luminosity LHC do? What about a 100 TeV pp collider?
    • What is the value added of the ILC? (CLIC? A high-luminosity circular e+e- collider?)
    • If deviations from SM Higgs couplings are detected can one extract a value for the mass scale of the new physics Ξ›?
    • How reliable is the determination of Ξ›, and how is this quantity related to a measurable quantity?
    • How many standard deviations are required for the deviations to be convincing? (e.g., how well do you need to measure the triple Higgs coupling?)
  • What is the fate of the Higgs self-coupling Ξ»(Q) as Qβ†’M_{PL}?
    • Is the Higgs vacuum stable or metastable?
    • What is the theoretical origin of Ξ»?
    • How does BSM physics impact these questions?
      • For example, in the MSSM, Ξ» is determined by gauge couplings, and the Higgs vacuum is therefore stable.
      • In other BSM models, the corresponding answers may not be so straightforward.
  • Is the gauge hierarchy problem resolved by TeV-scale physics? If yes, does this new physics provide us with a more fundamental understanding
    of the origin of electroweak symmetry breaking?
    • TeV-scale supersymmetrry remains a favored candidate for addressing the gauge hierarchy problem.
    • The SUSY wishlist for Higgs physics includes:
      • A resolution to the ΞΌ problem.
      • A more accurate computation of the Higgs mass to reduce the uncertainty below 1 GeV.

In addition, specific questions were directed to the ATLAS and CMS experimentalists.

My comments in red (Pierre Savard) (also EG=Eilam Gross)
  • Is the Ξ³Ξ³ excess statistically significant?
    • no

  • Do the ZZ*β†’ 4 lepton events provide a consistent story (relative to Ξ³Ξ³)?
    • yes, after Run 1

  • Are the ATLAS and CMS Higgs data self-consistent?
    • yes, after Run 1

  • How much tension is there with the SM expectations?
    • Not much (see ATLAS/CMS combination CONF note). ~10% compatibility. ttH (driven my multi-leptons) is high.
    • [HG] bb is also slightly low
  • We are eager for some clarifications…
    • Can custodial symmetry in the Higgs couplings be verified (ultimately with a similar accuracy to the ρ-parameter)?
    • Can we experimentally verify that fermion masses arise from the same mechanism as the gauge boson masses?
    • Is BR(h β†’ Ξ³Z) consistent with BR(h β†’ Ξ³Ξ³)?
    • What is BR(h β†’ non-SM channels)?
    • What is BR(h β†’ invisble)?
      • Current upper limit at 95% CL is about 20%

  • Seeking further clarifications…
    • Confirm spin and CP quantum numbers of the boson.
      • Alternative spin hypotheses very strongly disfavoured as is a pure CP-odd state

    • Measure the htt coupling (better yet: h coupling to the top partners, if they exist!)
      • Current significance is close to 4.5 sigmas in the combination
    • Eventually, measure the hhh coupling.
    • Try to detect the WWhh coupling, and identify potential BSM physics effects in the gghh box diagram.
    • Detect the WWWW quartic vertex and make sure that the gauge structure is preserved.
  • Beyond the SM Higgs boson---more wishes...
    • Find the charged Higgs boson.
    • (EG) Theorists need to sort out the overlap region when the charged Higgs mass is around the mass of the top so there is a production via t-->H+b and tH+b
    • Measure tan Ξ² (if you are absolutely certain that the Higgs sector corresponds to a Type-I or II 2HDM).
    • Even better---if you suspect that the Higgs sector corresponds to a 2HDM, measure the basis-independent Yukawa coupling matrices, and experimentally determine the structure of the Higgs-fermion coupling (which may not be of Type I or II).
    • Are there two nearly mass-degenerate scalars with mass around 125 GeV?

      • The compatibility of the data with the presence of a single state is discussed in upcoming ATLAS/CMS combination paper
    • (EG) The talk we heard Wednesday morning on the gg fusion calculation was excellent, need to clarify the linear vs quadrature. In my opinion it is difficult for the experiments to do the linear (technically) and also to do a rectangular error. We know how to do it but I believe it is extremely difficult to automate it, if possible at all,

April 18, 11:20 am, Bill Murray: Update for LFV Higgs searches

  • H-> mu+tau_e released in plot form; less sensitivity from ATLAS wrt CMS
  • does not confirm nor does it rule out the CMS excess.

April 18, 11:30 am, Christophe Grojean: EFT-4-Higgs

  • goal to establish a consistent framework to study deviations
  • answer the question of how do we go beyond the mu and kappa framework
  • discussed the pros and cons of the EFT approach
  • not much work on non-linear Higgs EFTs. At leading order very close to kappa framework.
  • no unique choice for a basis
  • simplifications are needed because of the limitation of current data, i.e. reduce the number of parameters
  • how many parameters have not been probed before the Higgs discovery: 8
  • 6 have been constrained with the current data (~15%), 2 outstanding. There is a flat direction which will be resolved by ttH
  • experimentally we don’t have access to the cut-off scale because the couplings and cut-off scale are β€œmeasured” together
  • EFT interpretation fail once the scales above the cut-off scale are included in the constraints
  • (Eilam Gross) I will be happy to have a clear guidance of the Kappa framework vs EFT, just to make clear that we should pursue the Kappa framework analysis.

April 18, 2:00 pm, Nathaniel Craig: Discussion on 750 GeV diphoton excess

List of Questions and comments:

  • List of basics questions from Nathaniel:
    • Are we happy with the current analysis? [yes]
    • If the signal is readl, did we have to get lucky twice? [no] 8TeV vs 13 TeV: diphotons kills diphotons? [no]
    • Wide or narrow: still not clear, but see discussion by B. Murray, below
    • Is it a resonance or a cascade? latest Moriond releases strongly disfavor cascade.
    • Is it a Higgs (i.e. does it contribute to the fermion/gauge boson mass?)
      • Higgs-singlet mixing strongly disfavored by WW
      • pure 2HDM (no extra stuff in the loops): very unlikely because of ttbar
      • 2HDM could be accommodated in larger framework.
    • Who ordered that?
      • susy: sneutrino (has the right quantum numbers) in RPV models, sgoldstino (not usual one though), sbino in supersoft susy/Dirac gauginos
      • composite: maybe in non-minimal models SO(6)/SO(4)
      • KK graviton: not in minimal model (with SM on the brane) because of dilepton channels. What about other models? see Hewett/Rizzo, Falkowski/Kamenik
      • radion is doing Ok, according to Gunion.
      • [added by P. Langacker: I would just like to emphasize that vector exotics (with masses stabilized by new symmetries) and SM singlets occur frequently in string constructions, and they are even required by tadpole consistency in many string embeddings of the MSSM. That was emphasized by Cvetic and Halverson and I several years ago, and was the main point in our 750 papers. Like other motivations for vectorlike exotics, the excess can be easily described if it is narrow.]
    • What's next? gammaZ seems a safe bet.
  • Question regarding the tail of the distribution: does the tail affect the fit? (answer: impact small as the fit is dominated by high stats region at low mass). Other question: do the models consider a bump + tail (answer: no)?
  • Question regarding the theory prediction: does this assume no gamma-jet background? Answer: do not know but the contamination is small (about 10%, going down vs mass)
  • Public info on narrow width with ATLAS 8 TeV (only 6% width results are reported). Answer: no public information
  • Comment: we should also see the bump in gamma, gamma-star. Other comment: ok, but by then, the yy bump will have been clearly established
  • Follow-up to Bill’s study: what is the probability of reconstructing a width that is greater than experimental resolution if the true width is narrow?
  • Question on jj: in the graviton interpretation, the sensitivity is 3 orders less
    • Decay to two photons that are boosted, reconstruction as a single photon β€”> are the photons single photons? can look at conversion probability. Also, in many models decay length is another handle
  • calibration 750 photon: what is th precision?
    • Answer from ATLAS (from note): ...detailed checks of the validity of the calibration for the lowest gain range of the electronics readout [38] of the electromagnetic calorimeter, which is used in the ET range above 350 GeV in the central part of the electromagnetic barrel calorimeter, have been performed, including checks with high-ET electrons from Z-boson decays. These checks show that the relative calibration of the low-gain readout with respect to the other gains is better than Β±1%.
  • dijets are sensitive in many models (but from graviton example above they are less sensitive)


April 19, 11:00 am, John Gunion: Discussion of extended Higgs sectors: 2HDM

Comments and questions
  • Getting rid of the wrong sign hbb coupling branch in the tan 𝛽 vs. cos(𝛽-𝛼) plane in Type II Models. The recent analysis by Misiak et al. that improves the calculation of the decay rate for b β†’ s+𝛾 in the SM and the 2HDM implies that the charged Higgs mass of the Type II 2HDM must be heavier than 480 GeV. Taking this limit along with constraints from the T parameter (which limits the size of the mass difference of H^+ and A) apparently kills a significant fraction of the wrong sign hbb branch previously allowed by the LHC Higgs data. (Some authors have in fact claimed that the branch is entirely killed off). If would be useful to clarify the situation.
  • Requests by theorists to extend searches for scalars decaying into 𝜏+𝜏- or into 𝛾𝛾 in the mass range below 100 GeV. There seems to already be searches for 𝜏+𝜏- down to 25 GeV in mass and 𝛾𝛾 down to 60 GeV in mass. (ATLAS/CMS experimentalists should confirm these numbers)
  • In 2HDM scans, there is a non-zero cross-section times branching ratio of a 750 GeV H or A to decay into 𝛾𝛾, but the rate is typically too small by a factor of about 100. The best chance for a 2HDM explanation of the 750 GeV diphoton resonance is gg β†’ H β†’ AA β†’ 𝛾𝛾𝛾𝛾, where the two photons from the A decay are unresolved. Here, the A must lie below charmed quark threshold so that its allowed decays are into πœ‡+πœ‡-, e+e- and 𝛾𝛾. In 2HDM scans, it seems possible to make this work for m_H=200 GeV only if the scalar potential parameter |πœ†_5| is larger than 2πœ‹, although values of |πœ†_5| below 4πœ‹ appear to work. One must also check to see whether the A decay leads to a displace vertex (the A should decay before it reaches the electromagnetic calorimeter.
  • What should experimentalists focus on in Runs 2 and 3 in the search for 2HDM phenomena?
    • Continue vigorous search for H,A β†’ 𝜏+𝜏-
    • A β†’ Zh is suppressed in the alignment limit, but still could provide a viable signature
    • A limited mass region for the light CP-odd A is currently ruled out, but there is still a significant region that could yield a signal.
    • So far, there are limited phenomenological studies of CP-violating observables arising from potentially complex 2HDM scalar potential parameters. This phenomenology needs to be pursued more systematically.
  • Radion interpretation of 750 GeV diphoton resonance.
    • In a Randall-Sundrum type model, the KK-graviton and the radion provide two possible candidates for the diphoton resonance at 750 GeV.
    • Considering a model where the gluon resides in the bulk, the KK excitation of the gluon should be above 3 TeV in light of LHC bounds.
    • The KK-graviton is heavier than the KK-gluon (in the model under consideration, it is about 1.5 times heavier). Thus, focus on the possibility of a 750 GeV radion.
    • The Higgs boson, h(125) and the radion mix. The mixing parameter, πœ‰, is a priori a free parameter, although πœ‰=1/6 corresponds to a conformal limit of the theory. (Of course, the conformal symmetry is explicitly broken by the SM Higgs potential!)
    • Remarkably, in the model under consideration, for values of πœ‰ near 0.16, the couplings of the radion to tt and bb vanish, the coupling to hh almost vanishes and the couplings to WW and ZZ are suppressed (these couplings would vanish for slightly lower values of πœ‰). Meanwhile, the couplings of the radion to 𝛾𝛾 and to gg (g=gluon) remain signficant.
    • Thus, one can find preferred regions of the parameter space in which the cross section times branching ratio for gg β†’ πœ™ β†’ 𝛾𝛾 (where πœ™ is the radion) is in the range observed by the ATLAS/CMS diphoton excesses. Meanwhile, the cross section times branching ratio for gg β†’ πœ™ β†’ gg is found to be about 3 pb, below (but not too far below) the current ATLAS/CMS bounds. It was suggested that the ATLAS/CMS bounds could be improved by implementing some quark/gluon discrimination, since presumably the main 2 jet background at the LHC in the kinematical regime of the 750 GeV diphoton excess comes from final state quark-gluon dijets.
    • It appears that the partial width of πœ™ β†’ gg is maximal for values of πœ‰ near 1/6. Why is this true?
    • Reading the fine print, the model under consideration requires that k/M_{PL} should be greater than 2, where k is the curvature parameter. This is somewhat distressing, since the effective field theory approach to the radion would seem to rely on the perturbativity of the gravity sector, which in turn requires that k/M_{PL} should be somewhat less than 1. (A comfortable value for this ratio would be more like 0.1.) In particular, in the case of a perturbative gravity sector, the radion would actually be the lightest scalar of the theory, i.e. with mass below that of h(125).


April 19, 2:00 pm, Shinya Kanemura: Discussion of extended Higgs sectors: Beyond the 2HDM

Comments and questions
  • Various BSM motivations for extended Higgs sectors: (i) origin of electroweak symmetry breaking; (ii) baryogenesis; (iii) dark mattter; (iv) neutrino mass
  • Can experimentally test the extended Higgs sector in three ways: (i) indirectly via the precision h(125) program; (ii) direct evidence for new scalar states; and (iii) the hhh coupling is very sensitive to the dynamics of the extended Higgs sector.
  • Fingerprinting the extended Higgs sector via the experimental πœ…_V vs. πœ…_F plane. This study relies on a tree-level analysis of Higgs couplings. Note that πœ…_V > 1 is possible in the Georgi-Machacek (G-M) model and in the doublet-septet model. Both these models yield 𝜌=1 at tree level. In the G-M model, 𝜌 is actually infinitely renormalized, since the special conditions satisfied by the tree-level scalar potential are spoiled by custodial-breaking effects due to the hypercharge gauge interactions and the t-b mass difference. (Is 𝜌=1 a "natural relation" of the doublet-septet model, meaning that radiative corrections to 𝜌=1 are finite as they are in the SM Higgs theory and the 2HDM?)
  • Fingerprinting the extended Higgs sector is still viable even when one-loop electroweak radiative corrections are taken into account. One can discriminate among various extended Higgs sector models with precision Higgs data from the HL-LHC.
  • The doubly charged Higgs boson H++. Appears in the G-M model. Motivated by the Type-II seesaw for neutrino masses. There are some limits from ATLAS/CMS, e.g.http://arxiv.org/abs/1410.6315
  • The Type-II seesaw mechinism. Couple triplet field to neutrino bilinear. The triplet field couples with coupling strength πœ‡ to two neutral doublet fields, both of which are replaced by their vevs. The latter leads to a triplet vev of V=πœ‡v^2/M^2, where v=246 GeV and M is the triplet mass. Then the mass matrix of the neutrinos is hV, where h is the neutrino-triplet scalar Yukawa coupling matrix. In order to get observed nuetrino masses, we must have |h|<<1. Thus, e.g. the decay of the doubly charged triplet to 𝜏+𝜏+ would be suppressed.
  • An alternative scheme for obtaining neutrino masses---the Zee-Babu model which employs a doubley charged scalar singlet. In some cases, the decay of he doubly charged triplet to 𝜏+𝜏+ can be experimentally observed. One could then distinguish these two models, since in the Type-II seesaw, the taus are both left-handed but in the Zee-Baby model, the taus are both right-handed. These can be distinguished in the 𝜏+ β†’ πœ‹+𝜈 decay by examining the energy distribution of the final state pions.
  • The H^+ W^- Z vertex is a sensitive probe of exotic Higgs sectors (even when 𝜌=1 as in the G-M model).
  • The hhh coupling is very sensitive to the dynamics of the extended Higgs sector due to two possible effects: (i) tree-level Higgs mixing effects; and (ii) one-loop electroweak corrections which can be significantly larger than the naive expectation. This in turn has important implications for the nature of the electroweak phase transition. In particular, one may end up with a strongly first order phase transition, which is one of the necessary ingredients for a successful explanation of baryogenesis via the electroweak phase transition in the early universe.
    • In the SM, the one-loop correction to the hhh coupling goes like 3m_h^2/v [ 1- N_c m_t^4 / (3 πœ‹^2 v^2 m_h^2]. Note the large radiative correction proportional to m_t^4. In the 2HDM, the corresponding result is has an additional positive term inside the brackets that does like (M_{Phi}^2-M^2)^3 / (12 πœ‹^2 m_h^2 M_{Phi}^2), where M_{Phi}^2 = M^2 +πœ†_i v^2. Although this term formally decouples for M^2>>v^2, in the region where M is a few hunderd GeV, this correction can be very large. Scanning over the parameter space, one finds corrections that are typically 100% (and in some extreme regions can be corrections of 500%). These large radiative corrections are correlated with the strength of the first order electroweak phase transition, necessary for electroweak baryogenesis. By the way, one does not need such large corrections to the hhh vertex to have successful electroweak baryogenesis. Regions of the 2HDM parameter space with 20--50% corrections to the hhh vertex be consistent with electroweak baryogenesis.
    • In cases where the hhh coupling deviates significantly from its SM value due to tree-level Higgs mixing effects, one also sees a crrelation with the strength of the first order electroweak phase transition.
    • In the 2HDM, the large deviations in the hhh due to the radiative corrections noted above appears to always be a positive shift from the SM value of the hhh coupling.
    • It was suggested that it might be more useful to provide a complete 1-loop electroweak radiative correction to hh production at the LHC in the SM and in extended Higgs models. Of course, the corrections to the hhh couplings is part of the story here, but it is not the whole story.
  • In a number of the studies mentioned above, unitarity and stability conditions (e.g. the scalar potential should be bounded from below) are imposed at the tree-level. Can one extend these conditions to the one-loop level? If yes, how would the correspondig conditions at one-loop effect the parameter scans performed in the various extended Higgs sector studies?


April 20, 4:00 pm, Stefania Gori (SG): Discussion of exotic Higgs signals

Comments and questions
  • For consideration: new decay modes of the h(125)---(i) rare decays to SM particles; (ii) decays to new particles that are outside the SM. For the latter, consider either prompt decays or decays with displaced vertices; and consider decays with no missing energy or with missing energy.
  • Not considered in this presentation are Higgs decays that exhibit lepton flavor violation such as h β†’ πœ‡+𝜏-.
  • Not considered in this presentation is the very rare Higgs decay h β†’ e+e-.
  • Regarding rare Higgs decays to SM particles, three classes of decays were highlighted:
    • Final state of the form V𝛾, where V= πœ™, J/πœ“, 𝛢(1s), 𝛢(2s), 𝛢(3s), πœ”, 𝜌.
    • Final states: W^- K^+, W^- D^+_s, Z𝛢(1s), Z𝛢(2s), Z𝛢(3s), Zπœ”, Z𝜌.
    • Flavor-violating Higgs decays, with final states of the form M^*𝛾, where M=B_s, B_d, D_s, D, K and M^* is the corresponding spin-1 resonance.
  • Concerning the flavor-violating decays, the constraints from low-energy flavor observables imply that Higgs branching ratios into such final states are much too small to be ever seen at the LHC. Additionally, it is probably impossible to trigger on such decays at the LHC.
    Question: can you do better with more inclusive modes, such as h β†’ bs?
  • The current bounds on h β†’ J/πœ“ 𝛾 are about 500 times the SM prediction and yield a bound on πœ…_c of about 220. But this bound does not take into account other implications of such a ridiculously large Higgs coupling to charmed quarks. For example, the charmed loop contribution to gg β†’ h would be very enhanced and competative with the top quark loop. Indeed, just using the bounds on the Higgs width (of order a few GeV from direct measurements) reduces the upper bound on πœ…_c to around 100. By the way, extrapolating the h β†’ J/πœ“ 𝛾 search to the HL LHC would eventually yield abound on πœ…_c of about 90.
  • Can one do better by searching for h β†’ cc by improving charm tagging techniques at the LHC. Efforts are underway, but the current status at ATLAS and CMS is not public, so no details on the current sensitivity are available.
  • The current bounds on h β†’ 𝛢(1s) 𝛾 are about 10^5 times the SM prediction. If this could be reduced eventually to, say 100 times the SM prediction, then one would be able to distinguish between πœ…_b=1 and πœ…_b= -1. This is possible due to the fact that h β†’ 𝛢(1s) 𝛾 can arise via two mechanisms, one that involves directly the coupling of h to bb and the second which arises when one of the photons in h β†’ 𝛾𝛾 is off-shell and transforms into 𝛢(1s) [vector meson dominance]. Thus, the amplitude for h β†’ 𝛢(1s) 𝛾 is proportional to a linear combination of πœ…_b and πœ…_𝛾. The relevant coefficients are such that one can overcome the factor of 100 in distinguishing between πœ…_b=1 and πœ…_b= -1, by measuring the sign of πœ…_b/πœ…_𝛾.
  • It is even more challenging to try to measure πœ…_s at the LHC. BR(h β†’ πœ™ 𝛾) is simply too small. SG: Experimentally, would a search for this decay mode be feasible? Signature: one photon with sizable pT and two tracks (phi-> K+ K-) in opposite direction with respect to the photon.
  • Question: Can LHCb contribute? Answer: very unlikely since they will not have enough Higgs data. Their integrated luminosity is at least a factor of 10 lower than the integrated luminosity collected by ATLAS and CMS, and of course, LHCb is not a 4πœ‹ detector.
  • Next for consideration are Higgs decays to new particles that are outside the SM. Focus on the decay to "dark particles" that are singlets with respect to the SM gauge group. Simplest example: a singlet scalar s that couple to h(125) via the Higgs portal interaction |H|^2 s^2
  • A sensible goal: detect branching ratios (or set bounds) of order 10%.
  • More generally, one can build simplified models based on three different portals: the Higgs portal (e.g. |H|^2 s^2), gauge kinetic mixing of the hypercharge field strength tensor with the field strength tensor of a "dark Z" or "dark photon", and the right-handed neutrino interaction HLN.
  • Concerning resonant searches with no missing transverse energy, e.g. Higgs decays into a dark Z_D/pseudoscalar, where the Z_D/pseudoscalar can decay into bb, 𝜏+𝜏-, πœ‡+πœ‡-, or e+e-, questions for experientalists:
    • How viable are the bbbb and bb𝜏+𝜏- signatures?
    • SG: recommendation: perform searches in a as broad as possible mass range.
    • How useful are the VBF + additional soft particle triggers for more background limited signatures?
    • Would it be useful to formulate additional benchmark models that combine several possible searches? (Bill Murray was skeptical).
    • What are the experimental prospects for these Higgs decays channels?
    • Better benchmark models are needed here.
    • If a signal is seen, how do we know if it has anything to do with h(125)?
  • Concerning displaced Higgs decays. Examples arise in neutral naturalness models (e.g. twin Higgs, fraternal twin Higgs, etc.). Existing methods focus on two displaced objects or one displaced object along with a high-threshold associated object. Experimental searches so far focus on displaced vertices at least 1 mm--1 cm from the collision point.
    • Can one extend the above method to the case of one displaced objject along with thresholds suitable for Higgs decays?
    • Can one improve the displace vertex analysis so as to be sensitive to displaced vertices that are as close as 50 πœ‡m from the collision point?
    • Remark: displaced signatures are more problematical at the HL LHC (due to pile-up issues).
  • General remark on search for exotic phenomena: specific benchmark models that focus on a single signature are useful for motivating experimental searches.

April 21, 4:00 pm, Pre-Close-out discussion: updating and extending the Higgs wishlist.
Some comments from Heather Gray [HG]
Contribution from Christophe Grojean:
  • Is the Higgs solitary?
  • Is the EW sector natural?
  • Is the Higgs playing any role in cosmology (inflation, baryogenesis, DM portal…)?
  • Is cosmology playing any role for the Higgs (e.g. relaxation of the weak scale via a coupling to an axion)?
  • Is the Higgs composite or elementary?
  • Is the Higgs responsible for flavor?
  • Is the Higgs breaking CP?
  • Is there a connection between the Higgs and the (possible) diphoton resonance at 750 GeV?
  • Are the masses of all fermions (in particular the ones of the light generations) proportional to the Higgs vev?
  • Are there any vector-like fermions with mass not originating from the Higgs vev?
Contribution from Sabine Kraml:
- How well can we measure/exploit VH production?
- Higgs-to-Higgs decays as priority signatures of extended Higgs sectors in Run 2:
  • A generic search for a heavy resonance decaying into a lighter one plus a Z-boson. The signature will be ll+X, where thell comes from the Z -> ll decay and X=bb,tautau, .... This was already done by CMS in Run 1 but there is no such ATLAS analysis (ATLAS considered only A -> Zh with mH fixed at 125 GeV). Interpretations include
    • 2HDM with mh=125 GeV : H -> ZA
    • 2HDM with mH=125 GeV : A -> ZH and A -> Zh over the full H, h mass ranges.
  • Heavy resonances decaying into a pair of lighter ones: H -> hh, where h is the SM-like 125 GeV Higgs; H -> AA, where A is a (pseudo-)scalar that can lie anywhere between mH/2 and very low mass; h -> AA; how low can one go in mA?

Contribution from Pierre Savard (using some inputs from the 2012 wishlist
  • Regarding the 125 GeV Higgs
    • establish H->bb decay:
      • [HG] Unless we're extremely unlucky this should come during Run-2 of the LHC.
      • [HG] The main limiting factor is likely to be modelling uncertainties on the V+bb backgrounds. More accurate predictions for these will directly impact the analyses
    • establish ttH production (currently high in the multi-lepton final states in particular)
      • [HG] Similarly, here we would benefit a lot from improved background estimates.
    • Measure a second generation coupling (H->mumu). Can it be done in Run 2 combining CMS and ATLAS?

      • [HG] Determine what limits the (HL)-LHC will be able to place on the H->cc couplings
  • How well can the two experiments do on H->Zgamma in Run 2?
  • Continue to push limits on H->invisible (currently a bit above 20%)
  • Continue program of precision measurements of sigma x BR in established channels. Any deviations?
    • reporting of results: kappa framework, template xs, differential xs, pseudo-observables, EFTs -> need to agree on benchmarks for combinations
    • [HG] Are there particular differential distributions that would be most useful to measure?
  • Is the observable Higgs state at 125 GeV responsible for the unitarization of W_L W_L scattering?
  • Continue program of measuring CP properties of 125 Higgs: spin is done, Pure CP-odd is excluded, any evidence of CP-mixing?
  • Detect WWWW quartic vertex and make sure that the gauge structure is preserved

April 21, 4:00 pm, Close-out discussion: updating and extending the Higgs wishlist.
We started the discussion from the remarks by Christophe and contributions by Pierre, Sabine, and Heather and produced the following wishlist
  • Variations of the question whether the Higgs sector is already fully discovered.
    • Is there an extended Higgs sector?
    • Are there new colorless scalars to be discovered?
    • Is there a Higgs portal?
  • Up to which scale is the Higgs elementary (not composite)?
  • Is there flavor violation mediated by the Higgs?
    • Can the constraints from the H(125) be reevaluated?
  • Is the Higgs breaking CP?
    • Are there additional channels?
    • Can we evaluate the complementarity to EDM experiments?
  • Is there a connection between the Higgs and the (possible) diphoton resonance at 750 GeV?
    • Can we probe a H^2phi^2 coupling?
  • Are the masses of all fermions (in particular the ones of the light generations) proportional to the Higgs vev?
    • How well can we assess the charm coupling via Higgs production?
  • Are there any vector-like fermions with their mass not originating from the Higgs vev?
  • Where can we enhance the signatures tested at the LHC
    • How well can we explore differential (like pT) distributions to find or to constrain new physics?
    • H_2 to H_1 to H_0, i.e. allow the asymmetric decay
    • extend the mass reach up, down, and intermediate for searches
    • additional signature for non-minimal extended Higgs sectors
      • like in LR models or other examples
  • What can be done experimentally in the next 3 years?
    • establish H->bb decay
    • establish ttH
    • second generation coupling (H->mumu)
  • Evaluate the Higgs data with EFT’s but focus on thoses which break degeneracies
  • Improvement on MC tools are needed to reduce uncertainties in background estimation. A comprehensive list of use cases would be helpful
    • examples are Z+jets, ttbar+jet
  • Re-evaluate the ultimate LHC performance for coupling measurements
    • add evaluation of H to charm coupling





April 15, 2:00 pm, Kaustubh Agashe: Discussions on strongly coupled BSM Higgs