MadGraph5_aMC@NLO

Suspicious zero x-section SMEFTsim model

Asked by Giacomo

Dear MadGraph experts,

I write to report an unexpected behaviour.
To give you a little bit of background, i am studying the effect of dimension 6 EFT operators to vector boson scattering (VBS) processes. In order to simulate all the Wilson coefficient-dependent terms (The SM interference, the pure-BSM quadratic one and the interference between pairs of operators) i am using the SMEFTsim model.

All the simulations went fine for the "one operator case" for the processes of same sign WW production, opposite sign WW EWK, inclusive WW and opposite sign WW QCD production.

Regarding the generation of the interference terms between two operators, MadGraph is not able to compute the amplitudes for all the combinations involving the operator named cHDD for the opposite sign WW QCD production [1]. Restriction cards are created putting every non-interesting operator to exactly 0 while leaving the two interesting ones at 9.999999e-01 [2].

To summarise some important point and to point out why this is suspicious:

1) Mg5 computes non zero amplitudes for the interference term between the SM and the cHDD for all the VBS processes.
2) Mg5 computes non zero amplitudes for the interference term between cHDD and every other interesting operator for all the VBS processes except for OSWW QCD production.
3) Mg5 computes non zero amplitudes for the interference term between every operators pair excluding cHDD for the OSWW QCD production.

A similar problem was found in this question: https://answers.launchpad.net/mg5amcnlo/+question/679764

After a long discussion with one of the SMEFTsim developer (Ilaria Brivio) we are still puzzled about this behaviour that may be due to madrgraph internal computations.
I suppose that this is a know problem from the answers to the aforementioned launchpad question so what could one do in order to obtain a reliable simulation in you opinion?

Thank you for your time.
Best regards,
Giacomo

-------------------------------------------------------
[1]
Here i report the madgraph commands issued in order to generate events for the opposite sign WW QCD production:

import model SMEFTsim_U35_MwScheme_UFO_v3_0-cHDD_cHWB_massless

generate p p > e+ ve mu- vm~ j j QCD==2 SMHLOOP=0 NPprop=0 NP==1 NPcHDD^2==1 NPcHWB^2==1

add process p p > e- ve~ mu+ vm j j QCD==2 SMHLOOP=0 NPprop=0 NP==1 NPcHDD^2==1 NPcHWB^2==1

Output MadGraph:

import /afs/cern.ch/work/g/gboldrin/public/public/CMSSW_10_2_13/src/D6EFTStudies/generation/launch_OSWWQCD_cHDD_cHWB_IN.txt
The import format was not given, so we guess it as command
import model SMEFTsim_U35_MwScheme_UFO_v3_0-cHDD_cHWB_massless
INFO: Restrict model SMEFTsim_U35_MwScheme_UFO_v3_0 with file ../MG5_aMC_v2_6_5/models/SMEFTsim_U35_MwScheme_UFO_v3_0/restrict_cHDD_cHWB_massless.dat .
INFO: Run "set stdout_level DEBUG" before import for more information.
INFO: Change particles name to pass to MG5 convention
Defined multiparticle p = g u c d s u~ c~ d~ s~
Defined multiparticle j = g u c d s u~ c~ d~ s~
Defined multiparticle l+ = e+ mu+
Defined multiparticle l- = e- mu-
Defined multiparticle vl = ve vm vt
Defined multiparticle vl~ = ve~ vm~ vt~
Defined multiparticle all = g u c d s u~ c~ d~ s~ a ve vm vt e- mu- ta- ve~ vm~ vt~ e+ mu+ ta+ t b t1 t~ b~ t1~ z w+ z1 w1+ h h1 w- w1-
generate p p > e+ ve mu- vm~ j j QCD==2 SMHLOOP=0 NPprop=0 NP==1 NPcHDD^2==1 NPcHWB^2==1
INFO: Trying process: g g > e+ ve mu- vm~ g g NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
INFO: Trying process: g g > e+ ve mu- vm~ u u~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
INFO: Trying process: g g > e+ ve mu- vm~ u c~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
INFO: Trying process: g g > e+ ve mu- vm~ c u~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
INFO: Trying process: g g > e+ ve mu- vm~ c c~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
INFO: Trying process: g g > e+ ve mu- vm~ d d~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
INFO: Trying process: g g > e+ ve mu- vm~ d s~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
INFO: Trying process: g g > e+ ve mu- vm~ s d~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
…

Command "import /afs/cern.ch/work/g/gboldrin/public/public/CMSSW_10_2_13/src/D6EFTStudies/generation/launch_OSWWQCD_cHDD_cHWB_IN.txt" interrupted in sub-command:
"generate p p > e+ ve mu- vm~ j j QCD==2 SMHLOOP=0 NPprop=0 NP==1 NPcHDD^2==1 NPcHWB^2==1" with error:
NoDiagramException : No amplitudes generated from process Process: g/u/c/d/s/u~/c~/d~/s~ g/u/c/d/s/u~/c~/d~/s~ > e+ ve mu- vm~ g/u/c/d/s/u~/c~/d~/s~ g/u/c/d/s/u~/c~/d~/s~ NP=1 NPprop=0 QCD=2 SMHLOOP=0 NP==1 QCD==2 NP^2==2 NPcHDD^2==1 NPcHWB^2==1 QCD^2==4 @1. Please enter a valid process
command not executed: add process p p > e- ve~ mu+ vm j j QCD==2 SMHLOOP=0 NPprop=0 NP==1 NPcHDD^2==1 NPcHWB^2==1
command not executed: output OSWWQCD_cHDD_cHWB_IN_v2

[2]
A snippet from the restriction card used to generate the interference between two operators for a pathological case

…
## INFORMATION FOR SMEFTCUTOFF
###################################
Block SMEFTcutoff
    1 1.000000e+03 # LambdaSMEFT

###################################
## INFORMATION FOR SMEFT
###################################
Block SMEFT
   1 0 # cG
   2 0 # cW
   3 0 # cH
   4 0 # cHbox
   5 9.999999e-01 # cHDD
   6 0 # cHG
   7 0 # cHW
   8 0 # cHB
   9 9.999999e-01 # cHWB
   10 0 # ceHRe
   11 0 # cuHRe
   12 0 # cdHRe
   13 0 # ceWRe
   14 0 # ceBRe
   15 0 # cuGRe
   16 0 # cuWRe
   17 0 # cuBRe
   18 0 # cdGRe
   19 0 # cdWRe
   20 0 # cdBRe
   21 0 # cHl1
   22 0 # cHl3
   23 0 # cHe
   24 0 # cHq1
   25 0 # cHq3
   26 0 # cHu
   27 0 # cHd
   28 0 # cHudRe
   29 0 # cll
   30 0 # cll1
   31 0 # cqq1
   32 0 # cqq11
   33 0 # cqq3
   34 0 # cqq31
   35 0 # clq1
   36 0 # clq3
   37 0 # cee
   38 0 # cuu
   39 0 # cuu1
   40 0 # cdd
   41 0 # cdd1
   42 0 # ceu
   43 0 # ced
   44 0 # cud1
   45 0 # cud8
   46 0 # cle
   47 0 # clu
   48 0 # cld
   49 0 # cqe
   50 0 # cqu1
   51 0 # cqu8
   52 0 # cqd1
   53 0 # cqd8
   54 0 # cledqRe
   55 0 # cquqd1Re
   56 0 # cquqd11Re
   57 0 # cquqd8Re
   58 0 # cquqd81Re
   59 0 # clequ1Re
   60 0 # clequ3Re
###################################

## INFORMATION FOR SMEFTCPV

###################################

Block SMEFTcpv
   1 0 # cGtil
   2 0 # cWtil
   3 0 # cHGtil
   4 0 # cHWtil
   5 0 # cHBtil
   6 0 # cHWBtil
   7 0 # ceWIm
   8 0 # ceBIm
   9 0 # cuGIm
   10 0 # cuWIm
   11 0 # cuBIm
   12 0 # cdGIm
   13 0 # cdWIm
   14 0 # cdBIm
   15 0 # cHudIm
   16 0 # ceHIm
   17 0 # cuHIm
   18 0 # cdHIm
   19 0 # cledqIm
   20 0 # cquqd1Im
   21 0 # cquqd8Im
   22 0 # cquqd11Im
   23 0 # cquqd81Im
   24 0 # clequ1Im
   25 0 # clequ3Im
…

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Olivier Mattelaer
Solved:
Last query:
Last reply:
Revision history for this message
Olivier Mattelaer (olivier-mattelaer) said : 		#1

Hi,

Do I understand correctly that your issue is that MG5aMC does not find any valid diagram?
I just want to be sure because this is coompletly differrent from a case where diagram exists and then the integration over the phase-space returns a zero cross-section (which is the first impression when reading your title/description and the bug report that you link).

If the above is correct, this is likely because the UFO model does not allow us to generate such diagram.
likely due some value within couplings_orders.py related to the max value of NP.
Your model is not part of the UFO model database (mainly all the UFO model put on the FeynRules website) so I can not be more precise.

Cheers,

Olivier

> On 28 Dec 2020, at 15:45, Giacomo <email address hidden> wrote:
>
> New question #694707 on MadGraph5_aMC@NLO:
> https://answers.launchpad.net/mg5amcnlo/+question/694707
>
> Dear MadGraph experts,
>
> I write to report an unexpected behaviour.
> To give you a little bit of background, i am studying the effect of dimension 6 EFT operators to vector boson scattering (VBS) processes. In order to simulate all the Wilson coefficient-dependent terms (The SM interference, the pure-BSM quadratic one and the interference between pairs of operators) i am using the SMEFTsim model.
>
> All the simulations went fine for the "one operator case" for the processes of same sign WW production, opposite sign WW EWK, inclusive WW and opposite sign WW QCD production.
>
> Regarding the generation of the interference terms between two operators, MadGraph is not able to compute the amplitudes for all the combinations involving the operator named cHDD for the opposite sign WW QCD production [1]. Restriction cards are created putting every non-interesting operator to exactly 0 while leaving the two interesting ones at 9.999999e-01 [2].
>
> To summarise some important point and to point out why this is suspicious:
>
> 1) Mg5 computes non zero amplitudes for the interference term between the SM and the cHDD for all the VBS processes.
> 2) Mg5 computes non zero amplitudes for the interference term between cHDD and every other interesting operator for all the VBS processes except for OSWW QCD production.
> 3) Mg5 computes non zero amplitudes for the interference term between every operators pair excluding cHDD for the OSWW QCD production.
>
> A similar problem was found in this question: https://answers.launchpad.net/mg5amcnlo/+question/679764
>
>
> After a long discussion with one of the SMEFTsim developer (Ilaria Brivio) we are still puzzled about this behaviour that may be due to madrgraph internal computations.
> I suppose that this is a know problem from the answers to the aforementioned launchpad question so what could one do in order to obtain a reliable simulation in you opinion?
>
> Thank you for your time.
> Best regards,
> Giacomo
>
> -------------------------------------------------------
> [1]
> Here i report the madgraph commands issued in order to generate events for the opposite sign WW QCD production:
>
> import model SMEFTsim_U35_MwScheme_UFO_v3_0-cHDD_cHWB_massless
>
> generate p p > e+ ve mu- vm~ j j QCD==2 SMHLOOP=0 NPprop=0 NP==1 NPcHDD^2==1 NPcHWB^2==1
>
> add process p p > e- ve~ mu+ vm j j QCD==2 SMHLOOP=0 NPprop=0 NP==1 NPcHDD^2==1 NPcHWB^2==1
>
>
> Output MadGraph:
>
> import /afs/cern.ch/work/g/gboldrin/public/public/CMSSW_10_2_13/src/D6EFTStudies/generation/launch_OSWWQCD_cHDD_cHWB_IN.txt
> The import format was not given, so we guess it as command
> import model SMEFTsim_U35_MwScheme_UFO_v3_0-cHDD_cHWB_massless
> INFO: Restrict model SMEFTsim_U35_MwScheme_UFO_v3_0 with file ../MG5_aMC_v2_6_5/models/SMEFTsim_U35_MwScheme_UFO_v3_0/restrict_cHDD_cHWB_massless.dat .
> INFO: Run "set stdout_level DEBUG" before import for more information.
> INFO: Change particles name to pass to MG5 convention
> Defined multiparticle p = g u c d s u~ c~ d~ s~
> Defined multiparticle j = g u c d s u~ c~ d~ s~
> Defined multiparticle l+ = e+ mu+
> Defined multiparticle l- = e- mu-
> Defined multiparticle vl = ve vm vt
> Defined multiparticle vl~ = ve~ vm~ vt~
> Defined multiparticle all = g u c d s u~ c~ d~ s~ a ve vm vt e- mu- ta- ve~ vm~ vt~ e+ mu+ ta+ t b t1 t~ b~ t1~ z w+ z1 w1+ h h1 w- w1-
> generate p p > e+ ve mu- vm~ j j QCD==2 SMHLOOP=0 NPprop=0 NP==1 NPcHDD^2==1 NPcHWB^2==1
> INFO: Trying process: g g > e+ ve mu- vm~ g g NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
> INFO: Trying process: g g > e+ ve mu- vm~ u u~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
> INFO: Trying process: g g > e+ ve mu- vm~ u c~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
> INFO: Trying process: g g > e+ ve mu- vm~ c u~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
> INFO: Trying process: g g > e+ ve mu- vm~ c c~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
> INFO: Trying process: g g > e+ ve mu- vm~ d d~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
> INFO: Trying process: g g > e+ ve mu- vm~ d s~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
> INFO: Trying process: g g > e+ ve mu- vm~ s d~ NPprop=0 SMHLOOP=0 NP==1 QCD==2 NPcHDD^2==1 NPcHWB^2==1 @1
> …
>
> Command "import /afs/cern.ch/work/g/gboldrin/public/public/CMSSW_10_2_13/src/D6EFTStudies/generation/launch_OSWWQCD_cHDD_cHWB_IN.txt" interrupted in sub-command:
> "generate p p > e+ ve mu- vm~ j j QCD==2 SMHLOOP=0 NPprop=0 NP==1 NPcHDD^2==1 NPcHWB^2==1" with error:
> NoDiagramException : No amplitudes generated from process Process: g/u/c/d/s/u~/c~/d~/s~ g/u/c/d/s/u~/c~/d~/s~ > e+ ve mu- vm~ g/u/c/d/s/u~/c~/d~/s~ g/u/c/d/s/u~/c~/d~/s~ NP=1 NPprop=0 QCD=2 SMHLOOP=0 NP==1 QCD==2 NP^2==2 NPcHDD^2==1 NPcHWB^2==1 QCD^2==4 @1. Please enter a valid process
> command not executed: add process p p > e- ve~ mu+ vm j j QCD==2 SMHLOOP=0 NPprop=0 NP==1 NPcHDD^2==1 NPcHWB^2==1
> command not executed: output OSWWQCD_cHDD_cHWB_IN_v2
>
>
>
> [2]
> A snippet from the restriction card used to generate the interference between two operators for a pathological case
>
> …
> ## INFORMATION FOR SMEFTCUTOFF
> ###################################
> Block SMEFTcutoff
> 1 1.000000e+03 # LambdaSMEFT
>
> ###################################
> ## INFORMATION FOR SMEFT
> ###################################
> Block SMEFT
> 1 0 # cG
> 2 0 # cW
> 3 0 # cH
> 4 0 # cHbox
> 5 9.999999e-01 # cHDD
> 6 0 # cHG
> 7 0 # cHW
> 8 0 # cHB
> 9 9.999999e-01 # cHWB
> 10 0 # ceHRe
> 11 0 # cuHRe
> 12 0 # cdHRe
> 13 0 # ceWRe
> 14 0 # ceBRe
> 15 0 # cuGRe
> 16 0 # cuWRe
> 17 0 # cuBRe
> 18 0 # cdGRe
> 19 0 # cdWRe
> 20 0 # cdBRe
> 21 0 # cHl1
> 22 0 # cHl3
> 23 0 # cHe
> 24 0 # cHq1
> 25 0 # cHq3
> 26 0 # cHu
> 27 0 # cHd
> 28 0 # cHudRe
> 29 0 # cll
> 30 0 # cll1
> 31 0 # cqq1
> 32 0 # cqq11
> 33 0 # cqq3
> 34 0 # cqq31
> 35 0 # clq1
> 36 0 # clq3
> 37 0 # cee
> 38 0 # cuu
> 39 0 # cuu1
> 40 0 # cdd
> 41 0 # cdd1
> 42 0 # ceu
> 43 0 # ced
> 44 0 # cud1
> 45 0 # cud8
> 46 0 # cle
> 47 0 # clu
> 48 0 # cld
> 49 0 # cqe
> 50 0 # cqu1
> 51 0 # cqu8
> 52 0 # cqd1
> 53 0 # cqd8
> 54 0 # cledqRe
> 55 0 # cquqd1Re
> 56 0 # cquqd11Re
> 57 0 # cquqd8Re
> 58 0 # cquqd81Re
> 59 0 # clequ1Re
> 60 0 # clequ3Re
> ###################################
>
> ## INFORMATION FOR SMEFTCPV
>
> ###################################
>
> Block SMEFTcpv
> 1 0 # cGtil
> 2 0 # cWtil
> 3 0 # cHGtil
> 4 0 # cHWtil
> 5 0 # cHBtil
> 6 0 # cHWBtil
> 7 0 # ceWIm
> 8 0 # ceBIm
> 9 0 # cuGIm
> 10 0 # cuWIm
> 11 0 # cuBIm
> 12 0 # cdGIm
> 13 0 # cdWIm
> 14 0 # cdBIm
> 15 0 # cHudIm
> 16 0 # ceHIm
> 17 0 # cuHIm
> 18 0 # cdHIm
> 19 0 # cledqIm
> 20 0 # cquqd1Im
> 21 0 # cquqd8Im
> 22 0 # cquqd11Im
> 23 0 # cquqd81Im
> 24 0 # clequ1Im
> 25 0 # clequ3Im
> …
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
Giacomo (giacbold) said : 		#2

Hi,

Yes you are right, MG5 does not find any diagram for that sintax. A zero-cross section is obtained when generating without restrictions, where MG5 finds some diagrams, and then restricting to 0 non interesting operators.

From other tests we can already say that the interference is compatible with 0, but i think that some diagrams should enter in the computation. I will take a closer look at the couplings definition.

May i ask you what do you mean by "...related to the max value of NP." ?

You can find the SMEFTsim model here, if needed: https://feynrules.irmp.ucl.ac.be/wiki/SMEFT

Thank you for your answer,
Best regards,

Giacomo

Revision history for this message
Best Olivier Mattelaer (olivier-mattelaer) said : 		#3

> May i ask you what do you mean by "...related to the max value of NP." ?

you have a file coupling_orders.py
where you can specify that a given coupling value can not be higher than X.
This is the flag expansion_order = 99
#99 means no restriction

> You can find the SMEFTsim model here, if needed:
> https://feynrules.irmp.ucl.ac.be/wiki/SMEFT

I found the version 2.3 but not the version 3.0.
I have even run the automatic detection script to be sure that it was not added since the last time.

This is the list of all model that I know that start by SMEFTsim
(either on FR website or on my laptop)

SMEFTsim_A_general_alphaScheme_UFO_v2 SMEFTsim_A_general_MwScheme_UFO_v2/ SMEFTsim_A_MFV_MwScheme_UFO_v2_1 SMEFTsim_A_U35_alphaScheme_UFO_v2_1/ SMEFTsim_A_U35_MwScheme_UFO_v2_1 SMEFTsim_tb1/
SMEFTsim_A_general_MwScheme_UFO_v2 SMEFTsim_A_MFV_alphaScheme_UFO_v2_1 SMEFTsim_A_U35_alphaScheme_UFO_v2_1 SMEFTsim_A_U35_MwScheme_UFO/ SMEFTsim_A_U35_MwScheme_UFO_v2_1/ SMEFTsim_top_test/

Cheers,

Olivier

> On 28 Dec 2020, at 17:11, Giacomo <email address hidden> wrote:
>
> Question #694707 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/694707
>
> Status: Answered => Open
>
> Giacomo is still having a problem:
> Hi,
>
> Yes you are right, MG5 does not find any diagram for that sintax. A
> zero-cross section is obtained when generating without restrictions,
> where MG5 finds some diagrams, and then restricting to 0 non interesting
> operators.
>
>> From other tests we can already say that the interference is compatible
> with 0, but i think that some diagrams should enter in the computation.
> I will take a closer look at the couplings definition.
>
> May i ask you what do you mean by "...related to the max value of NP." ?
>
> You can find the SMEFTsim model here, if needed:
> https://feynrules.irmp.ucl.ac.be/wiki/SMEFT
>
> Thank you for your answer,
> Best regards,
>
> Giacomo
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.

Revision history for this message
Giacomo (giacbold) said : 		#4

Thanks Olivier Mattelaer, that solved my question.