ICHEP Preparations

June 15, 2012

In just three weeks time ICHEP 2012 will be underway with the biggest expected news to come from Higgs Searches presented on 9th July in Melbourne. Meanwhile closer to home a much more low-key meeting at CERN has given us an update on the running of the LHC. As far as I can tell from the images of the auditorium the only people who attend these meetings in person are the speakers, but there are high quality webcasts so nobody else has to.

The usual LHC Machine Status Report by Steve Myers is worth watching if you are interested in where they stand with beam operations. Tonight they will have a celebration for collecting 5/fb although they have already passed the 6/fb mark. This merged slide shows progress as we approach the next technical stop compared to his prediction from Chamonix. It is good news that they are almost on target and will decide not to use the two months of extra running that had been set aside just in case. They will need the extra time to do the growing pile of tasks scheduled for the long shutdown which could easily extend to nearly two years long if they are not careful. If you are observant you will also notice that the main goal is illustrated with a picture I took for this blog last year but Myres will have to nick a lot more of my stuff if he is to get even for all the ones I took from him 🙂

Myers also reported a scary story about a situation that nearly led to LHC armagedon last week when they discovered in testing that the beam dump system relied on a power supply that formed a single point of failure. A simple fault could have led to a situation where the beams could not be dumped even when a failure signal would normally abort the run. The beams would have kept circulating like an unstoppable train with the only possible outcome being the loss of 120 MJ of beam energy around the accelerator ring with the potential to destroy almost anything and everything in the collider.

The progress talk from ATLAS is also interesting in that they revealed three new versions of Higgs plots using 2011 data with improved analysis methods. From this I deduce that at ICHEP they will update only the critical digamma and Z to 4l channels with 2012 data and will use these new versions of the 2011 data for some of the other channels. These will be combined to form the new ATLAS plot. Although the cut-off for collecting data for ICHEP has passed I think there is a good chance they will continue collecting up to the technical stop because the collection rate is now very high. This may give them close to 6/fb of data in the two high-resolution channels.

In the ICHEP abstracts CMS indicate that they will update all their channels with 2012 data. This difference between the two experiments is similar to what was presented at the December council meeting.  As far as I know there is no technical reason why the two experiments should not combine all the 2011 data and 2012 data in the high-resolution channels in time for ICHEP, but this would risk arriving at the unfair situation where one of the two experiments gets the discovery level significance first. I think there is a good chance that they will compare preliminary results in the next few days and if it is clear that they can both reach the critical 5 sigma level they may go for full combinations and announce the joint discovery. Whether they can do this depends on how much data they use, how well they deal with pile-up, how well they control the background, how lucky they are with the fluctuations and even how big the cross-section is if the Higgs is seeing BSM enhancements. My prediction is that they wont quite get there this time round, however an unofficial discovery using ATLAS+CMS is much more likely.

Apart from these Higgs results there will be a lot of other new results presented at ICHEP including a few exotic searches (such as heavy gauge bosons) using 2012 data from both ATLAS and CMS. I expect ATLAS to hold their 2012 supersymmetry searches back for SUSY 2012 in August but the abstracts from CMS indicate that a few new SUSY at 8 TeV will be shown at ICHEP. LHCb, Tevatron and many others from the accelerator labs, neutrino experiments and astronomical observatories have the potential to produce new discoveries in particle physics and ICHEP 2012 is the place to grab the headlines so we should expect the unexpected.

If you have not yet voted in the 3 Quarks Daily science blog awards viXra log would appreciate your support. Update: voting is complete and we appear to have made it through to the next round in about 9th place with 68 votes. Thanks.

Bayes and String Theory

June 12, 2012

If Supersymmetry is found or excluded at the Large hadron Collider, how will it affect your opinion on string theory as unification of gravity and particle physics? This is a hard question and opinions differ widely across the range of theorists, but at the least any answer should be consistent with the laws of probability including Bayes Law. What can we really say?

A staunch string theorist might want to respond as follows:

“I am confident about the relevance of superstring theory to the unification of gravity and the forces of elementary particles because it provides a unique way to accomplish this that is consistent in the perturbative limits (Amongst other reasons.) Unfortunately it does not have a unique solution for the vacuum and we have not yet found a principle for selecting the solution that applies to our universe. Because of this we cannot predict the low energy effective physics and we cannot even know if supersymmetry is an observable feature of physics at energy scales currently accessible. Therefore if supersymmetry is not observed at the TeV scale even after the LHC has explored all channels up to 14 TeV with high integrated luminosities, there is no reason for that to make me doubt string theory. On the other hand, if supersymmetry is observed I will be enormously encouraged. This is because there are good reasons to think that supersymmetry will be restored as an exact gauge symmetry at some higher scale, and gauged sypersymmetry inevitably includes gravity within some version of supergravity. There are further good reasons why supergravity is not likely to be fully consistent on its own and would necessarily be completed only as a limit of superstring theory. Therefore if supersymmetry is discovered by the LHC my confidence in string theory will be greatly improved.” 

On hearing this a string theory skeptic would surely be seen shaking his head vigorously. He would say:

“You cannot have it both ways! If you believe that the discovery of supersymmetry will confirm string theory then you must also accept that failure to discover it falsify string theory. Any link between the two must work equally in both directions. You are free to say that supersymmetry at the electro-weak scale is a theory completely Independent of string theory if you wish. In that case you are safe if suppersymmetry is not found but by the same rule the discovery of supersymmetry cannot be used to claim that superstring theory is right. If you prefer you can claim that superstring theory predicts supersymmetry (some string theorists do) but if that is your position you must also accept that excluding supersymmetry at the LHC will mean that string theory has failed. You can take a position in between but it must work equally in both directions.”

  The Tetrahedron of Possibilities

What does probability theory tell us about the range of possibilities that a theorist can consider for answers to this problem? Prior to the experimental result he will have some estimate for the probability that string theory is a correct theory of quantum gravity and for the probability that supersymmetry will be observed at the LHC. In my case I assign a probability of PST = 0.9 to the idea that string theory is correct and PSUSY = 0.7 to the probability that SUSY will be seen at the LHC. These are my prior probabilities based on my knowledge and reasoning. You can have different values for your estimates because you know different things, but you can’t argue with mine. There are no absolutely correct global values for these probabilities, they are a relative concept.

However, these two probabilities do not describe everything I need to know. There are four logical outcomes I need to consider altogether:

  • P1 = the probability that both string theory is correct and SUSY will be found
  • P2 = the probability that string theory is correct and SUSY will not be found
  • P3 = the probability that string theory is wrong and SUSY will be found
  • P4 = the probability that string theory is wrong and SUSY will not be found

You might try to tell me that there are other possibilities, such as that SUSY exists at higher energies or that string theory is somehow partly right, but I could define my conditions for correctness of string theory and for discovery of SUSY so that they are unambiguous. I will assume that has been done. This means that the four possible outcomes are mutually exclusive and exhaustive. We can conclude that P1 + P2 + P3 + P4 = 1. Of course the four probabilities must also be between 0 and 1. These conditions map out a three-dimensional tetrahedron in the four-dimensional space of the four probability variables with the four logical outcomes at each vertex. This is the tetrahedron of possible prior probabilities and any theorists prior assessment of the situation must be described by a single point within this tetrahedron.

So far I have only given two values that describe my own assessment so to pinpoint my complete position within the three-dimensional range I must give one more value. If I thought that string theory and SUSY at the weak scale were completely independent theories I could just multiply as follows

P1 = PST .PSUSY = 0.63
P2 = PST .(1 – PSUSY) = 0.27
P3 = (1 – PST) .PSUSY = 0.07
P4 = (1 – PST) .(1 – PSUSY) = 0.03

The condition that the two theories are independent fall on a surface given by the equation P1 . P4 = P2 . P3 that neatly divides the tetrahedron in two.

As I already explained I do not think these two things are independent. I think that SUSY would strongly imply string theory. In other words I think that the probability of SUSY being found and string theory being wrong is much lower than the value of 0.07 for P3 . In fact I estimate it to be something like P3 = 0.01. I must still keep the other probabilities fixed so P1 + P2 = PST = 0.9 and P1 + P3 = PSUSY = 0.7. This means that all my probabilities are now known

P1 = 0.69
P2 = 0.21
P3 = 0.01
P4 = 0.09

Notice that I did not get to fix P1 separately from P3. If I know how much the discovery of SUSY is going to affect my confidence in string theory then I also know how much the non-discovery of SUSY will affect it. It is starting to sound like the string theory skeptic could be right, but wait. Let’s see what happens after the LHC has finished looking.

Suppose SUSY is now discovered, how does this affect my confidence? My posterior probabilities P’2 and P’4 both become zero and by the rules of conditional probabilities P’ST = P1/PSUSY = 0.69/0.7 = 0.986. In other words my confidence in string theory will have jumped from 90% to 98.6%, quite a significant increase. But what happens if SUSY is found to be inaccessible to the LHC? In that case we end up with P’ST = P2/(1-PSUSY) = 0.21/0.3 = 0.7 . This means that my confidence in string theory will indeed be dented, but it is far from falsified. I should still consider string theory to have much better than level odds. So the skeptic is not right. The string theorist can argue that finding SUSY will be a good boost to string theory without it being falsified if SUSY is excluded, but the string theorists has to make a small concession too. His confidence in string theory has to be less if SUSY is not found.

Remember, I am not claiming that these probabilities are universally correct. They represent my assessment and I am not a fully fledged string theorist. Someone who has studied it more deeply may have a higher prior confidence in which case excluding SUSY will not make much difference at all to him even if he believes SUSY would strongly imply string theory.

LHC Prepares for ICHEP

June 10, 2012

According to an earlier commenter on this blog, today is the cut-off date for the LHC to collect data for the ICHEP conference.  Despite technical issues,  overall the LHC has been “enjoying remarkable availability” Right on cue the luminosity delivered has passed the 5/fb mark promised by the operations group. The total LHC delivered luminosity is now about 11.2/fb for CMS and will soon pass the 11.87/fb total for the Tevatron run II.

Now the computer grid will light up as ATLAS and CMS push through the analysis for the Higgs and SUSY plots in time to get them approved for the massive ICHEP conference in one months time.

The schedule of talks  shows that each experiment will be giving detailed talks for each individual Higgs channel (plus possibly a new H -> Z+γ analysis) using 2012 data at 8 TeV leading up to the final combinations for each experiment. They will leave it up to bloggers to complete the full combination. It is likely that they will fall just short of discovery significance in diphoton channels and combined plots for each experiment. The full combined significance for the LHC will probably pass the 5 sigma finish line but no official combination will show that. This is not certain because the statistical fluctuations are not predictable.

In the lead up to ICHEP there have been many lesser HEP conferences already but none have had data beyond 2011 available from the LHC. Even at this late stage ATLAS has added a new conference note with an update to the WW channel using multivariate analysis which improves the sensitivity from 2011 data. There have been a few interesting talks about the significance of the Higgs excess at 125 GeV and some comments about the fact that the diphoton channels are showing a stronger signal than expected while the WW channel is noticeably deficient. These irregularities are likely to be statistical but they are a sign of the interesting speculations that will follow further results.

Some questions we will be looking to answer:

  • How much will the excesses seen in 2011 data be strengthened?
  • Will the CMS and ATLAS peaks still be at slightly different masses?
  • Will the channel branching ratios remain significantly different from standard model predictions?
  • Will the 115GeV to 120 GeV window be excluded?
  • Will the DG continue to call the observations just “interesting fluctuations” and get away with it?

On a slightly different topic, the 3quarksdaily blog has launched its science blog competition for 2012 which will be judged by fellow blogger Sean Carroll. About a hundred shameless bloggers have nominated themselves for the title and voting will begin for the final cut in the next couple of days. I have nominated my coverage of the December Higgs announcement so please support viXra by voting in our favour. Update: link for voting is here