jump to navigation

Physics analysis vs blogging: 6-0 December 23, 2007

Posted by dorigo in personal, physics.

I have been blogging for three years now, and I enjoy it a lot – I find it stimulating to share my views and sometimes have to defend them, as well as to distribute information related to my job to peers, and of course to make my little bit of outreach and explain particle physics to laymen. I spend blogging overall more than one hour a day, which is an outrageously large amount of time, considering I could be accused of stealing it from my job and from my family. On the other hand, the truth is I am stealing this time from other lazy occupations I used to entertain myself with: chess, or bridge, or other internet activities of lesser value.

During these three intense years I have often found out that the urge of posting something interesting (at least to me) every day or so supersedes other pastimes and obligations. I almost always manage to find the time for a post. Almost.

In the last couple of weeks I have started a full immersion in a difficult analysis for the CMS experiment: attempting the extraction of a signal of associated production of a top quark pair and a Higgs boson, with top quarks decaying into hadrons and missing energy, and the Higgs decaying to a pair of b-quark jets. The analysis of course is based on simulated events, but it is a very important first step into understanding the extractability of the signal from a final state nobody has cared to study before. So, the issue is stimulating. Actually, the complexity of the task -the signal is buried in a background at least 10,000 times larger- makes it even more exciting.

So, during these last few weeks I have found out that when I am on to something like finding a signal or coding a smart algorithm for the search, I forget everything else, including my blog! I think this is a very healthy reaction which shows I have not spent all my cartridges yet as a physicist. Of course, I have had other exciting studies to delve into during the past three years, but I admit this last one is really absorbing all my intellectual energy. Right now is a Sunday evening, I am home, with my family, the christmas tree is a few feet from me, and here I am, finally finding ten minutes to write on my blog while a root macro is running five million events to produce a relative likelihood distribution…

It would be nice if I could discuss the analysis in more detail, but I have to go by the rules: even if the data I am analyzing is simulated, it still refers to the CMS detector and so I cannot divulgate anything of what I am finding. I can only say that there is good reason why associated top and Higgs production was taken off the list of golden signatures at the LHC (and the recent CMS Physics TDR in fact does not even bother discussing it): the signal is indeed darn tough to find… But I have my own aces up my sleeve, and I am confident I can prove that something could in principle be squeezed out of the very uncommon “jets + missing Et” top signature – the one that allowed me and my PhD student, Giorgio Cortiana, to publish the third-best top pair cross section measurement, three years ago.



1. Marco - December 24, 2007

Tanti auguri Tommaso! And good luck with your analysis 🙂

2. Michael Schmitt - December 24, 2007

Ciao Tommaso! Did you consider looking at higgs decays to W’s or to a pair of photons? The rates are lower but the signal much more distinctive than multiple jets. You probably aren’t planning to discover the Higgs with this messy signature, so what matters is confirming a discovery in a main channel by observations in related channels. And then you could go on to say something about top-Higgs Yukawa couplings, hopefully.

3. nc - December 24, 2007

“In the last couple of weeks I have started a full immersion in a difficult analysis for the CMS experiment: attempting the extraction of a signal of associated production of a top quark pair and a Higgs boson, with top quarks decaying into hadrons and missing energy, and the Higgs decaying to a pair of b-quark jets. The analysis of course is based on simulated events…”

Suppose the existing Higgs theories are wrong, and the electroweak sector of the standard model isn’t right. Maybe it needs the inclusion of gravity into the symmetry group, and maybe that upsets existing predictions. What do you do about including “null hypotheses” when you do statistical checks? I think all this experimental particle physics comes down to statistics and that means that there is a risk of getting the occasional good correlation from flawed theory which just happens to coincide with the experimental result.

One example was that Yukawa predicted a heavy meson (as the strong force mediator) in the 1930s and because the muon was soon discovered with about the right mass, it was assumed to be the field quanta. A decade later, after he won a Nobel Prize, it was found that the field quanta was the pion, not the muon. Wishful thinking is very possible with fuzzy data where there may be several different possible theoretical interpretations, only one of which is right. Sorry if this sounds an unhelpful comment. I enjoy reading your blog.

4. Andrea Giammanco - December 24, 2007

Tommaso: it would be certainly interesting for your readers to know why ttH was considered a golden channel until a few years ago (e.g. in the ATLAS TDR), and now it’s considered an incredibly challenging signature (e.g. in the CMS TDR).
As far as I know (please correct me if I’m wrong) this paradigm shift is mostly due to the advent of multi-parton matrix element generators like alpgen and madgraph, and of proper “matching” recipes, which permitted a more realistic modeling of the tt+Nj background.

Michael: we have also people in CMS working on these signatures, don’t worry:)
Since the final state is very different, it is impractical that the same people studying in detail the ttbb final state also find the time to accurately explore the ttgammagamma and ttWW final states. But this has been done in some specialized MC study in the last few years.

5. dorigo - December 25, 2007

Hi Marco, auguri di buone feste anche a te!

Yes, Michael, the idea is that once the signal is found in the WW or ZZ channel (or in the gamma-gamma channel if the Higgs mass is lowish), one wants to characterize it and determine its properties, to see if the beast we caught is exactly the one we were hunting for.

The tt-gamma gamma channel is being studied by others. But it is an idea when the tt is tagged by missing Et and jets, which is something others have neglected so far. However, the PhD thesis of my student who is working with me at the ttH study starts from a study of pixels in the trigger to reconstruct jets and discriminate a multijet topology at high luminosity – something you can do with a reconstruction of the vertex z from which jets come. In that case, the ttbb signature benefits more from the trigger enhancement, and so the thesis makes more sense with it.


6. dorigo - December 25, 2007

Hi Nigel,

particle physics has made some progress since the pi meson. Phenomenologists are on the lookout for signatures of unexpected physics, signature-based searches are on, and monte carlo simulations are tuned even too well.

I think you should not fear a mistaken interpretation. Only, what I think is that we will not see anything odd to interpret one way or another 🙂

Andrea, yes, a post on that issue would be interesting. Would you like to write it ? 😉
Ah, and – Michael is a member of CMS and he’s actually high up the ladder. He is into Higgs physics since more time than me and you, so his comment was didactical in nature rather than explorative.


7. Tumbledried - December 25, 2007

Merry christmas Tommaso! As an interested layman who knows very little about the specifics of experimental particle physics, I have little to add, but I do enjoy reading your few posts on these matters. It is great to hear that you are still getting plenty of new ideas! All the best for your research in the new year.

8. Andrea Giammanco - December 26, 2007

I apologize for having misunderstood the meaning of Michael’s comment, indeed his name sounds a bit familiar…
Anyway thanks for the invitation, but I’m not the most entitled person for talking about MC generators and/or matching recipes 🙂

9. Nikita Nikolaev - December 27, 2007

Good luck and keep us up to date as much as you feel you should. We’re eager, but we understand.

10. How many times do we have to discover the Higgs boson? « Collider Blog - December 24, 2008

[…] that collider physicists seem to want to find the Higgs over and over again… Some time ago, Tommaso Dorigo mounted a very challenging strategy of using the ttH channel in hadronic mode (ie, the Higgs is […]

Sorry comments are closed for this entry

%d bloggers like this: