MadGraph5_aMC@NLO

Higgs Characterization model problem

Asked by Florian Bernlochner

Hello,

I ran into some problems running the code from the Higgs Characterization model*. I contacted the authors, but I am not sure if it might be linked to a problem in MG5_aMC@NLO (?)

The proper NLO diagrams are generated, but once I try to produce events & integrate the cross section out I get

INFO: Checking test output:
INFO: P0_uu_x0uu
INFO: Result for test_ME:
INFO: Passed.
INFO: Result for test_MC:
INFO: Passed.
INFO: Result for check_poles:
Error detected in "launch auto "
write debug file /afs/cern.ch/work/f/florianb/sw/MG5_aMC_v2_1_1/bin/X0_jj_default/run_01_tag_1_debug.log
If you need help with this issue please contact us on https://answers.launchpad.net/madgraph5
aMCatNLOError : Poles do not cancel, run cannot continue

Thanks!
Florian

* = http://feynrules.irmp.ucl.ac.be/wiki/HiggsCharacterisation

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
Federico Demartin Edit question
Solved by:
Florian Bernlochner
Solved:
Last query:
Last reply:
Revision history for this message
Federico Demartin (federico-demartin) said : 		#1

Ciao Florian,

it's probably not a problem of the model, nor a problem of MG5_aMCatNLO, but a problem of the process;
I bet you have generated VBF... but this is just a guess based on past experience.

In order to help you, I need complete information on how you imported the model and generated the X0jj process, i.e. I need the full list of commands you prompted after opening madgraph.
It can also be helpful to know the settings you put in the param_card and in the run_card.
Please give me all this info. For the cards, you can just send me an email with those files attached.
Btw, I cannot open your link to run_01_tag_1_debug.log (maybe I need a cern account or other credentials that I don't have).

Cheers,

Federico

Revision history for this message
Florian Bernlochner (florian-bernlochner-0) said : 		#2

Hi Federico,

Thanks for the prompt reply and you are right :)

I did

MG5_aMC>import model HC_NLO_X0_UFO

instead of

MG5_aMC>import model HC_NLO_X0_UFO-heft

Now it works -- sorry for the noise and thanks again!

So, if I would want to produce VBF as well, will that problem pop up again?

Cheers,
Florian

Revision history for this message
Federico Demartin (federico-demartin) said : 		#3

Ciao Florian,

ah ok, so you were generating gluon fusion plus 2 jets, not VBF.
As you can see in the README file inside the model (which we ask to read carefully!), to generate GF you need to import the '-heft' model.
Instead, for the other relevant single higgs production channels (VBF, VH, ttH) you need to import the default model (4F scheme), or the '-no_b_mass' one (5F scheme).

The other day I guessed VBF because other people had problems with this process, and there are physical reasons behind this.
For example, problems can emerge from the "pentagon" diagrams that appear as NLO one-loop virtual corrections (imagine to add a t-channel gluon between the quark lines).
These diagrams are not pure QCD corrections (there are weak boson propagators inside the loop), so mg5_amcatnlo cannot compute them yet.
Anyway they can be safely neglected, their contribution is very small compared to the other one-loop corrections (= DIS approximation); but missing them can cause numerical miscancelation of the IR divergencies (poles in 1/epsilon in dimensional regularisation) and so madgraph complains.
In the README you find as well instructions to avoid this problem when generating VBF.
If other problems arise with VBF at NLO, just let me know.
No problem should arise when generating GF, VH or ttH, which have well-defined NLO corrections in QCD (I've extensively tested this before releasing the model).

Cheers,

Federico