As usual, we follow the first round of experimental talks with some post-coffee theory.
10:45am: Andreas Weiler, "Colliding Sflavours"
We generically expect top partners from naturalness. These correct the Higgs mass, and thus also affect the Higgs-gluon coupling. We can't directly probe this from gluon fusion of an s-channel Higgs, unless we have an additional hard jet. (This is precisely the idea that shows up in the paper I mentioned in my previous post.)
Now SUSY. Natural models with only the 3rd generation sfermions and the Higgsinos light offer some reason for optimism. Of course, we have direct stop searches that are placing non-trivial constraints on these possibilities. But depending on assumptions, stop can be down to a few hundred GeV.
One possible place to look for the effects of such a light stop is in the top production cross section. A stop with the top mass has a production cross section 17% as large; compared to theoretical errors on the top cross section at a few percent. This involves recasting the top cross section measurement. Limits can be drawn, in progress.
Naturalness wants split squarks, which faces tension with flavour constraints. How then can we achieve it? Partial alignments involves new physics flavour structure having some commonality with SM. For left-handed squarks this limits the splitting you can gain.
Model to get this involves gauge mediation with SU(3)F, the diagonal subgroup of the SM flavour symmetry. (That group is anomaly free.) Naturally splits third, first two generations; and generically lowers the third generation.
More generally, can have all squark split. This has a lot of effects on the standard limits, which assume degenerate squarks. Sea squarks (scharm, sstrange) can live at 400 GeV.
Project: FastLim, improve ability to see if models covered by experiments. Use extended basis of simplified models, to factorize the most costly CPU evaluation, of efficiencies and cross sections.
11:20am: Antonio Delgado, "Signatures of the least supersymmetric standard model"
This talk offers a description of a top-down construction of natural SUSY, and its phenomenology.
Amusing point: the very discovery of the Higgs makes things more fine-tuned; compare technicolour, for example. We now need to explain how the EW scale can be natural, without any new EW-scale particle.
The generic requirement for natural SUSY models is two sources of SUSY breaking; one for the third and one for the first two generations. Here, we have flavourful gauge mediation (squarks) combined with gravity mediation (stops). In particular, we charge the first two generations under a new U(1) and mediate SUSY breaking with that gauge group. The Higgses are uncharged under this gauge group, allowing them to gain their masses from the Giudice-Masiero mechanism. Meanwhile, gravity mediation for the third generation allows the creations of large A terms.
We need to break the U(1) to, among things, generate the Yukawa terms. This does lead to a small tuning between the VEV that breaks the U(1) and the cut-off scale that generates non-renormalisable operators. We also need to set off-diagonal elements small by hand, so this is not a theory of flavour.
This model ultimately gives us a TeV stop and 2 TeV gluino; note that we expect A ~ mstop, so we are not in the maximal mixing scenario. The electroweakinos are lighter, everything else is heavier. The dominant search comes from the coloured objects, to no surprise; these decay pretty much exactly as is assumed in simplified models, so there's not much new there. The analysis suggests that it might be observable at 14 TeV, though the limits fall rapidly with gluino mass.
11:55am: Nima Arkani-Hamed, "The Amplituhedron"
I wonder if this talk will be similar to the (over two hour) talk Nima gave at SUSY two years ago...
Yep, looks like it.
Is, however, claiming an important qualitative change in results in the last year.
Basic theme is an attempt to get a completely new way to analyse scattering physics that is much simpler. Analogy to classical to QM: minimising the classical action is not explicitly deterministic. This was a puzzle, but makes sense as the h goes to 0 limit of QM.
Strategy is to reformulate QFT with locality and unitarity as emergent phenomena.
Feynman diagrams make unitarity and locality manifest step by step. As a consequence they tell us about the structure of the poles, they only arise when certain sums of momenta become timelike.
A simpler representation comes in the cut structure of amplitudes. In planar theories with sufficiently much SUSY can take this further.
On-shell diagrams: alternate representation of scattering process in terms of purely on-shell particles, i.e. no virtual particles. Close relation to mathematical object, positive Grassmannian, in turn related to combinatorics. This in turn is related to an infinite-dimensional symmetry, the Yangian, that exists in scattering amplitudes but is obscured.
Problem was that the way to glue these building blocks together was still mandated by need for locality and unitarity.
Consider a triangle. Points inside triangle can be interpreted as centre of mass of varying masses at corners. The correct generalisation of this idea is precisely the positive Grassmannian. Consider polgyons. Given just the vertices, the only meaningful sense of inside/outside is if the polygon is convex. This in turn is related to the positive Grassmannian.
Duality between concept of area and a two-form with logarithmic singularities at vertices. This two-form is closely related to the concept of triangulation of the polygon.
Next generalisation is from polygons to polyhedrons. The triangulation in four-dimensions gives us the structure needed to explain unknown point mentioned above.
Now have a simple way to calculate large multiplicity scattering amplitudes in a geometric way that evades the length needed to do the Feynman calculation.
Loops are "high school geometry problem".
I do find this whole topic interesting and exciting, if quite difficult to follow. It's a bit hard to blog, because of the subtlety and because Nima talks quite fast.