What comes after the LHC? – The P5 Report & Future Colliders

This is the second part of our coverage of the P5 report and its implications for particle physics. To read the first part, click here

One of the thorniest questions in particle physics is ‘What comes after the LHC?’. This was one of the areas people were most uncertain what the P5 report would say. Globally, the field is trying to decide what to do once the LHC winds down in ~2040 While the LHC is scheduled to get an upgrade in the latter half of the decade and run until the end of the 2030’s, the field must start planning now for what comes next. For better or worse, big smash-y things seem to capture a lot of public interest, so the debate over what large collider project to build has gotten heated. Even Elon Musk is tweeting (X-ing?) memes about it.

Famously, the US’s last large accelerator project, the Superconducting Super Collider (SSC), was cancelled in the ’90s partway through its construction. The LHC’s construction itself often faced perilous funding situations, and required a CERN to make the unprecedented move of taking a loan to pay for its construction. So no one takes for granted that future large collider projects will ultimately come to fruition.

Desert or Discovery?

When debating what comes next, dashed hopes of LHC discoveries are top of mind. The LHC experiments were primarily designed to search for the Higgs boson, which they successfully found in 2012. However, many had predicted (perhaps over-confidently) it would also discover a slew of other particles, like those from supersymmetry or those heralding extra-dimensions of spacetime. These predictions stemmed from a favored principle of nature called ‘naturalness’ which argued additional particles nearby in energy to the Higgs were needed to keep its mass at a reasonable value. While there is still much LHC data to analyze, many searches for these particles have been performed so far and no signs of these particles have been seen.

These null results led to some soul-searching within particle physics. The motivations behind the ‘naturalness’ principle that said the Higgs had to be accompanied by other particles has been questioned within the field, and in New York Times op-eds.

No one questions that deep mysteries like the origins of dark matter, matter anti-matter asymmetry, and neutrino masses, remain. But with the Higgs filling in the last piece of the Standard Model, some worry that answers to these questions in the form of new particles may only exist at energy scales entirely out of the reach of human technology. If true, future colliders would have no hope of

A diagram of the particles of the Standard Model laid out as a function of energy. The LHC and other experiments have probed up to around 10^3 GeV, and found all the particles of the Standard Model. Some worry new particles may only exist at the extremely high energies of the Planck or GUT energy scales. This would imply a large large ‘desert’ in energy, many orders of magnitude in which no new particles exist. Figure adapted from here

The situation being faced now is qualitatively different than the pre-LHC era. Prior to the LHC turning on, ‘no lose theorems’, based on the mathematical consistency of the Standard Model, meant that it had to discover the Higgs or some other new particle like it. This made the justification for its construction as bullet-proof as one can get in science; a guaranteed Nobel prize discovery. But now with the last piece of the Standard Model filled in, there are no more free wins; guarantees of the Standard Model’s breakdown don’t occur until energy scales we would need solar-system sized colliders to probe. Now, like all other fields of science, we cannot predict what discoveries we may find with future collider experiments.

Still, optimists hope, and have their reasons to believe, that nature may not be so unkind as to hide its secrets behind walls so far outside our ability to climb. There are compelling models of dark matter that live just outside the energy reach of the LHC, and predict rates too low for direct detection experiments, but would be definitely discovered or ruled out by high energy colliders. The nature of the ‘phase transition’ that occurred in the very early universe, which may explain the prevalence of matter over anti-matter, can also be answered. There are also a slew of experimentalhints‘, all of which have significant question marks, but could point to new particles within the reach of a future collider.

Many also just advocate for building a future machine to study nature itself, with less emphasis on discovering new particles. They argue that even if we only further confirm the Standard Model, it is a worthwhile endeavor. Though we calculate Standard Model predictions for high energies, unless they are tested in a future collider we will not ‘know’ how if nature actually works like this until we test it in those regimes. They argue this is a fundamental part of the scientific process, and should not be abandoned so easily. Chief among the untested predictions are those surrounding the Higgs boson. The Higgs is a central somewhat mysterious piece of the Standard Model but is difficult to measure precisely in the noisy environment of the LHC. Future colliders would allow us to study it with much better precision, and verify whether it behaves as the Standard Model predicts or not.

Projects

These theoretical debates directly inform what colliders are being proposed and what their scientific case is.

Many are advocating for a “Higgs factory”, a collider of based on clean electron-positron collisions that could be used to study the Higgs in much more detail than the messy proton collisions of the LHC. Such a machine would be sensitive to subtle deviations of Higgs behavior from Standard Model predictions. Such deviations could come from the quantum effects of heavy, yet-undiscovered particles interacting with the Higgs. However, to determine what particles are causing those deviations, its likely one would need a new ‘discovery’ machine which has high enough energy to produce them.

Among the Higgs factory options are the International Linear Collider, a proposed 20km linear machine which would be hosted in Japan. ILC designs have been ‘ready to go’ for the last 10 years but the Japanese government has repeated waffled on whether to approve the project. Sitting in limbo for this long has led to many being pessimistic about the projects future, but certainly many in the global community would be ecstatic to work on such a machine if it was approved.

Designs for the ILC have been ready for nearly a decade, but its unclear if it will receive the greenlight from the Japanese government. Image source

Alternatively, some in the US have proposed building a linear collider based on a ‘cool copper’ cavities (C3) rather than the standard super conducting ones. These copper cavities can achieve more acceleration per meter than the standard super conducting ones, meaning a linear Higgs factory could be constructed with a reduced 8km footprint. A more compact design can significantly cut down on infrastructure costs that governments usually don’t like to use their science funding on. Advocates had proposed it as a cost-effective Higgs factory option, whose small footprint means it could potentially hosted in the US.

The Future-Circular-Collider (FCC), CERN’s successor to the LHC, would kill both birds with one extremely long stone. Similar to the progression from LEP to the LHC, this new proposed 90km collider would run as Higgs factory using electron-positron collisions starting in 2045 before eventually switching to a ~90 TeV proton-proton collider starting in ~2075.

An image of the proposed FCC overlayed on a map of the French/Swiss border
Designs for the massive 90km FCC ring surrounding Geneva

Such a machine would undoubtably answer many of the important questions in particle physics, however many have concerns about the huge infrastructure costs needed to dig such a massive tunnel and the extremely long timescale before direct discoveries could be made. Most of the current field would not be around 50 years from now to see what such a machine finds. The Future-Circular-Collider (FCC), CERN’s successor to the LHC, would kill both birds with one extremely long stone. Similar to the progression from LEP to the LHC, this new proposed 90km collider would run as Higgs factory using electron-positron collisions starting in 2045 before eventually switching to a ~90 TeV proton-proton collider starting in ~2075. Such a machine would undoubtably answer many of the important questions in particle physics, however many have concerns about the extremely long timescale before direct discoveries could be made. Most of the current field would not be around 50 years from now to see what such a machine finds. The FCC is also facing competition as Chinese physicists have proposed a very similar design (CEPC) which could potentially start construction much earlier.

During the snowmass process many in the US starting pushing for an ambitious alternative. They advocated a new type of machine that collides muons, the heavier cousin of electrons. A muon collider could reach the high energies of a discovery machine while also maintaining a clean environment that Higgs measurements can be performed in. However, muons are unstable, and collecting enough of them into formation to form a beam before they decay is a difficult task which has not been done before. The group of dedicated enthusiasts designed t-shirts and Twitter memes to capture the excitement of the community. While everyone agrees such a machine would be amazing, the key technologies necessary for such a collider are less developed than those of electron-positron and proton colliders. However, if the necessary technological hurdles could be overcome, such a machine could turn on decades before the planned proton-proton run of the FCC. It can also presents a much more compact design, at only 10km circumfrence, roughly three times smaller than the LHC. Advocates are particularly excited that this would allow it to be built within the site of Fermilab, the US’s flagship particle physics lab, which would represent a return to collider prominence for the US.

A proposed design for a muon collider. It relies on ambitious new technologies, but could potentially deliver similar physics to the FCC decades sooner and with a ten times smaller footprint. Source

Deliberation & Decision

This plethora of collider options, each coming with a very different vision of the field in 25 years time led to many contentious debates in the community. The extremely long timescales of these projects led to discussions of human lifespans, mortality and legacy being much more being much more prominent than usual scientific discourse.

Ultimately the P5 recommendation walked a fine line through these issues. Their most definitive decision was to recommend against a Higgs factor being hosted in the US, a significant blow to C3 advocates. The panel did recommend US support for any international Higgs factories which come to fruition, at a level ‘commensurate’ with US support for the LHC. What exactly ‘comensurate’ means in this context I’m sure will be debated in the coming years.

However, the big story to many was the panel’s endorsement of the muon collider’s vision. While recognizing the scientific hurdles that would need to be overcome, they called the possibility of muon collider hosted in the US a scientific ‘muon shot‘, that would reap huge gains. They therefore recommended funding for R&D towards they key technological hurdles that need to be addressed.

Because the situation is unclear on both the muon front and international Higgs factory plans, they recommended a follow up panel to convene later this decade when key aspects have clarified. While nothing was decided, many in the muon collider community took the report as a huge positive sign. While just a few years ago many dismissed talk of such a collider as fantastical, now a real path towards its construction has been laid down.

Hitoshi Murayama, chair of the P5 committee, cuts into a ‘Shoot for the Muon’ cake next to a smiling Lia Merminga, the director of Fermilab. Source

While the P5 report is only one step along the path to a future collider, it was an important one. Eyes will now turn towards reports from the different collider advocates. CERN’s FCC ‘feasibility study’, updates around the CEPC and, the International Muon Collider Collaboration detailed design report are all expected in the next few years. These reports will set up the showdown later this decade where concrete funding decisions will be made.

For those interested the full report as well as executive summaries of different areas can be found on the P5 website. Members of the US particle physics community are also encouraged to sign the petition endorsing the recommendations here.

The Search for Simplicity : The Higgs Boson’s Self Coupling

When students first learn quantum field theory, the mathematical language the underpins the behavior of elementary particles, they start with the simplest possible interaction you can write down : a particle with no spin and no charge scattering off another copy of itself. One then eventually moves on to the more complicated interactions that describe the behavior of fundamental particles of the Standard Model. They may quickly forget this simplified interaction as a unrealistic toy example, greatly simplified compared to the complexity the real world. Though most interactions that underpin particle physics are indeed quite a bit more complicated, nature does hold a special place for simplicity. This barebones interaction is predicted to occur in exactly one scenario : a Higgs boson scattering off itself. And one of the next big targets for particle physics is to try and observe it.

A feynman diagram consisting of two dotted lines coming merging together to form a single line.
A Feynman diagram of the simplest possible interaction in quantum field theory, a spin-zero particle interacting with itself.

The Higgs is the only particle without spin in the Standard Model, and the only one that doesn’t carry any type of charge. So even though particles such as gluons can interact with other gluons, its never two of the same kind of gluons (the two interacting gluons will always carry different color charges). The Higgs is the only one that can have this ‘simplest’ form of self-interaction. Prominent theorist Nima Arkani-Hamed has said that the thought of observing this “simplest possible interaction in nature gives [him] goosebumps“.

But more than being interesting for its simplicity, this self-interaction of the Higgs underlies a crucial piece of the Standard Model: the story of how particles got their mass. The Standard Model tells us that the reason all fundamental particles have mass is their interaction with the Higgs field. Every particle’s mass is proportional to the strength of the Higgs field. The fact that particles have any mass at all is tied to the fact that the lowest energy state of the Higgs field is at a non-zero value. According to the Standard Model, early in the universe’s history when the temperature were much higher, the Higgs potential had a different shape, with its lowest energy state at field value of zero. At this point all the particles we know about were massless. As the universe cooled the shape of the Higgs potential morphed into a ‘wine bottle’ shape, and the Higgs field moved into the new minimum at non-zero value where it sits today. The symmetry of the initial state, in which the Higgs was at the center of its potential, was ‘spontaneously broken’  as its new minimum, at a location away from the center, breaks the rotation symmetry of the potential. Spontaneous symmetry breaking is a very deep theoretical idea that shows up not just in particle physics but in exotic phases of matter as well (eg superconductors). 

A diagram showing the ‘unbroken’ Higgs potential in the very early universe (left) and the ‘wine bottle’ shape it has today (right). When the Higgs at the center of its potential it has a rotational symmetry, there are no preferred directions. But once it finds it new minimum that symmetry is broken. The Higgs now sits at a particular field value away from the center and a preferred direction exists in the system. 

This fantastical story of how particle’s gained their masses, one of the crown jewels of the Standard Model, has not yet been confirmed experimentally. So far we have studied the Higgs’s interactions with other particles, and started to confirm the story that it couples to particles in proportion to their mass. But to confirm this story of symmetry breaking we will to need to study the shape of the Higgs’s potential, which we can probe only through its self-interactions. Many theories of physics beyond the Standard Model, particularly those that attempt explain how the universe ended up with so much matter and very little anti-matter, predict modifications to the shape of this potential, further strengthening the importance of this measurement.

Unfortunately observing the Higgs interacting with itself and thus measuring the shape of its potential will be no easy feat. The key way to observe the Higgs’s self-interaction is to look for a single Higgs boson splitting into two. Unfortunately in the Standard Model additional processes that can produce two Higgs bosons quantum mechanically interfere with the Higgs self interaction process which produces two Higgs bosons, leading to a reduced production rate. It is expected that a Higgs boson scattering off itself occurs around 1000 times less often than the already rare processes which produce a single Higgs boson.  A few years ago it was projected that by the end of the LHC’s run (with 20 times more data collected than is available today), we may barely be able to observe the Higgs’s self-interaction by combining data from both the major experiments at the LHC (ATLAS and CMS).

Fortunately, thanks to sophisticated new data analysis techniques, LHC experimentalists are currently significantly outpacing the projected sensitivity. In particular, powerful new machine learning methods have allowed physicists to cut away background events mimicking the di-Higgs signal much more than was previously thought possible. Because each of the two Higgs bosons can decay in a variety of ways, the best sensitivity will be obtained by combining multiple different ‘channels’ targeting different decay modes. It is therefore going to take a village of experimentalists each working hard to improve the sensitivity in various different channels to produce the final measurement. However with the current data set, the sensitivity is still a factor of a few away from the Standard Model prediction. Any signs of this process are only expected to come after the LHC gets an upgrade to its collision rate a few years from now.

Limit plots on HH production in various different decay modes.
Current experimental limits on the simultaneous production of two Higgs bosons, a process sensitive to the Higgs’s self-interaction, from ATLAS (left) and CMS (right). The predicted rate from the Standard Model is shown in red in each plot while the current sensitivity is shown with the black lines. This process is searched for in a variety of different decay modes of the Higgs (various rows on each plot). The combined sensitivity across all decay modes for each experiment allows them currently to rule out the production of two Higgs bosons at 3-4 times the rate predicted by the Standard Model. With more data collected both experiments will gain sensitivity to the range predicted by the Standard Model.

While experimentalists will work as hard as they can to study this process at the LHC, to perform a precision measurement of it, and really confirm the ‘wine bottle’ shape of the potential, its likely a new collider will be needed. Studying this process in detail is one of the main motivations to build a new high energy collider, with the current leading candidates being an even bigger proton-proton collider to succeed the LHC or a new type of high energy muon collider.

Various pictorial representations of the uncertainty on the Higgs potential shape.
A depiction of our current uncertainty on the shape of the Higgs potential (center), our expected uncertainty at the end of the LHC (top right) and the projected uncertainty a new muon collider could achieve (bottom right). The Standard Model expectation is the tan line and the brown band shows the experimental uncertainty. Adapted from Nathaniel Craig’s talkhere

The quest to study nature’s simplest interaction will likely span several decades. But this long journey gives particle physicists a roadmap for the future, and a treasure worth traveling great lengths for.

Read More:

CERN Courier Interview with Nima Arkani-Hamed on the future of Particle Physics on the importance of the Higgs’s self-coupling

Wikipedia Article and Lecture Notes on Spontaneous symmetry breaking

Recent ATLAS Measurements of the Higgs Self Coupling

Measuring the Tau’s g-2 Too

Title : New physics and tau g2 using LHC heavy ion collisions

Authors: Lydia Beresford and Jesse Liu

Reference: https://arxiv.org/abs/1908.05180

Since April, particle physics has been going crazy with excitement over the recent announcement of the muon g-2 measurement which may be our first laboratory hint of physics beyond the Standard Model. The paper with the new measurement has racked up over 100 citations in the last month. Most of these papers are theorists proposing various models to try an explain the (controversial) discrepancy between the measured value of the muon’s magnetic moment and the Standard Model prediction. The sheer number of papers shows there are many many models that can explain the anomaly. So if the discrepancy is real,  we are going to need new measurements to whittle down the possibilities.

Given that the current deviation is in the magnetic moment of the muon, one very natural place to look next would be the magnetic moment of the tau lepton. The tau, like the muon, is a heavier cousin of the electron. It is the heaviest lepton, coming in at 1.78 GeV, around 17 times heavier than the muon. In many models of new physics that explain the muon anomaly the shift in the magnetic moment of a lepton is proportional to the mass of the lepton squared. This would explain why we are a seeing a discrepancy in the muon’s magnetic moment and not the electron (though there is a actually currently a small hint of a deviation for the electron too). This means the tau should be 280 times more sensitive than the muon to the new particles in these models. The trouble is that the tau has a much shorter lifetime than the muon, decaying away in just 10-13 seconds. This means that the techniques used to measure the muons magnetic moment, based on magnetic storage rings, won’t work for taus. 

Thats where this new paper comes in. It details a new technique to try and measure the tau’s magnetic moment using heavy ion collisions at the LHC. The technique is based on light-light collisions (previously covered on Particle Bites) where two nuclei emit photons that then interact to produce new particles. Though in classical electromagnetism light doesn’t interact with itself (the beam from two spotlights pass right through each other) at very high energies each photon can split into new particles, like a pair of tau leptons and then those particles can interact. Though the LHC normally collides protons, it also has runs colliding heavier nuclei like lead as well. Lead nuclei have more charge than protons so they emit high energy photons more often than protons and lead to more light-light collisions than protons. 

Light-light collisions which produce tau leptons provide a nice environment to study the interaction of the tau with the photon. A particles magnetic properties are determined by its interaction with photons so by studying these collisions you can measure the tau’s magnetic moment. 

However studying this process is be easier said than done. These light-light collisions are “Ultra Peripheral” because the lead nuclei are not colliding head on, and so the taus produced generally don’t have a large amount of momentum away from the beamline. This can make them hard to reconstruct in detectors which have been designed to measure particles from head on collisions which typically have much more momentum. Taus can decay in several different ways, but always produce at least 1 neutrino which will not be detected by the LHC experiments further reducing the amount of detectable momentum and meaning some information about the collision will lost. 

However one nice thing about these events is that they should be quite clean in the detector. Because the lead nuclei remain intact after emitting the photon, the taus won’t come along with the bunch of additional particles you often get in head on collisions. The level of background processes that could mimic this signal also seems to be relatively minimal. So if the experimental collaborations spend some effort in trying to optimize their reconstruction of low momentum taus, it seems very possible to perform a measurement like this in the near future at the LHC. 

The authors of this paper estimate that such a measurement with a the currently available amount of lead-lead collision data would already supersede the previous best measurement of the taus anomalous magnetic moment and further improvements could go much farther. Though the measurement of the tau’s magnetic moment would still be far less precise than that of the muon and electron, it could still reveal deviations from the Standard Model in realistic models of new physics. So given the recent discrepancy with the muon, the tau will be an exciting place to look next!

Read More:

An Anomalous Anomaly: The New Fermilab Muon g-2 Results

When light and light collide

Another Intriguing Hint of New Physics Involving Leptons

New detectors on the block

Article title: “Toward Machine Learning Optimization of Experimental Design”

Authors: MODE Collaboration

Reference: https://inspirehep.net/literature/1850892 (pdf)

In a previous post we wondered if (machine learning) algorithms can replace the entire simulation of detectors and reconstruction of particles. But meanwhile some experimentalists have gone one step further – and wondered if algorithms can design detectors.

Indeed, the MODE collaboration stands for Machine-learning Optimized Design of Experiments and in its first paper promises nothing less than that.

The idea here is that the choice of characteristics that an experiment can have is vast (think number of units, materials, geometry, dimensions and so on), but its ultimate goal can still be described by a single “utility function”. For instance, the precision of the measurement on specific data can be thought of as a utility function.

Then, the whole process that leads to obtaining that function can be decomposed into a number of conceptual blocks: normally there are incoming particles, which move through and interact with detectors, resulting in measurements; from them, the characteristics of the particles are reconstructed; these are eventually analyzed to get relevant useful quantities, the utility function among them. Ultimately, chaining together these blocks creates a pipeline that models the experiment from one end to the other.

Now, another central notion is differentiation or, rather, the ability to be differentiated; if all the components of this model are differentiable, then the gradient of the utility function can be calculated. This leads to the holy grail: finding its extreme values, i.e. optimize the experiment’s design as a function of its numerous components.

Before we see whether the components are indeed differentiable and how the gradient gets calculated, here is an example of this pipeline concept for a muon radiography detector.

Discovering a hidden space in the Great Pyramid by using muons. ( Financial Times)

Muons are not just the trendy star of particle physics (as of April 2021), but they also find application in scanning closed volumes and revealing details about the objects in them. And yes, the Great Pyramid has been muographed successfully.

In terms of the pipeline described above, a muon radiography device could be modeled in the following way: Muons from cosmic rays are generated in the form of 4-vectors. Those are fed to a fast-simulation of the scanned volume and the detector. The interactions of the particles with the materials and the resulting signals on the electronics are simulated. This output goes into a reconstruction module, which recreates muon tracks. From them, an information-extraction module calculates the density of the scanned material. It can also produce a loss function for the measurement, which here would be the target quantity.

Conceptual layout of the optimization pipeline. (MODE collaboration)

This whole ritual is a standard process in experimental work, although the steps are usually quite separate from one another. In the MODE concept, however, not only are they linked together but also run iteratively. The optimization of the detector design proceeds in steps and in each of them the parameters of the device are changed in the simulation. This affects directly the detector module and indirectly the downstream modules of the pipeline. The loop of modification and validation can be constrained appropriately to keep everything within realistic values, and also to make the most important consideration of all enter the game – that is of course cost and the constraints that it brings along.

Descending towards the minimum. (Dezhi Yu)

As mentioned above, the proposed optimization proceeds in steps by optimizing the parameters along the gradient of the utility function. The most famous incarnation of gradient-based optimization is gradient descent which is customarily used in neural networks. Gradient descent guides the network towards the minimum value of the error that it produces, through the possible “paths” of its parameters.

In the MODE proposal the optimization is achieved through automatic differentiation (AD), the latest word in the calculation of derivatives in computer programs. To shamefully paraphrase Wikipedia, AD exploits the fact that every computer program, no matter how complicated, executes a sequence of elementary arithmetic operations and functions. By applying the chain rule repeatedly to these operations, derivatives can be computed automatically, accurately and efficiently.

Also, something was mentioned above about whether the components of the pipeline are “indeed differentiable”. It turns out that one isn’t. This is the simulation of the processes during the passage of particles through the detector, which is stochastic by nature. However, machine learning can learn how to mimic it, take its place, and provide perfectly fine and differentiable modules. (The brave of heart can follow the link at the end to find out about local generative surrogates.)

This method of designing detectors might sound like a thought experiment on steroids. But the point of MODE is that it’s the realistic way to take full advantage of the current developments in computation. And maybe to feel like we have really entered the third century of particle experiments.

Further reading:

The MODE website: https://mode-collaboration.github.io/

A Beginner’s Guide to Differentiable Programming: https://wiki.pathmind.com/differentiableprogramming

Black-Box Optimization with Local Generative Surrogates: https://arxiv.org/abs/2002.04632

A symphony of data

Article title: “MUSiC: a model unspecific search for new physics in
proton-proton collisions at \sqrt{s} = 13 TeV”

Authors: The CMS Collaboration

Reference: https://arxiv.org/abs/2010.02984

First of all, let us take care of the spoilers: no new particles or phenomena have been found… Having taken this concern away, let us focus on the important concept behind MUSiC.

ATLAS and CMS, the two largest experiments using collisions at the LHC, are known as “general purpose experiments” for a good reason. They were built to look at a wide variety of physical processes and, up to now, each has checked dozens of proposed theoretical extensions of the Standard Model, in addition to checking the Model itself. However, in almost all cases their searches rely on definite theory predictions and focus on very specific combinations of particles and their kinematic properties. In this way, the experiments may still be far from utilizing their full potential. But now an algorithm named MUSiC is here to help.

MUSiC takes all events recorded by CMS that comprise of clean-cut particles and compares them against the expectations from the Standard Model, untethering itself from narrow definitions for the search conditions.

We should clarify here that an “event” is the result of an individual proton-proton collision (among the many happening each time the proton bunches cross), consisting of a bouquet of particles. First of all, MUSiC needs to work with events with particles that are well-recognized by the experiment’s detectors, to cut down on uncertainty. It must also use particles that are well-modeled, because it will rely on the comparison of data to simulation and, so, wants to be sure about the accuracy of the latter.

Display of an event with two muons at CMS. (Source: CMS experiment)

All this boils down to working with events with combinations of specific, but several, particles: electrons, muons, photons, hadronic jets from light-flavour (=up, down, strange) quarks or gluons and from bottom quarks, and deficits in the total transverse momentum (typically the signature of the uncatchable neutrinos or perhaps of unknown exotic particles). And to make things even more clean-cut, it keeps only events that include either an electron or a muon, both being well-understood characters.

These particles’ combinations result in hundreds of different “final states” caught by the detectors. However, they all correspond to only a dozen combos of particles created in the collisions according to the Standard Model, before some of them decay to lighter ones. For them, we know and simulate pretty well what we expect the experiment to measure.

MUSiC proceeded by comparing three kinematic quantities of these final states, as measured by CMS during the year 2016, to their simulated values. The three quantities of interest are the combined mass, combined transverse momentum and combined missing transverse momentum. It’s in their distributions that new particles would most probably show up, regardless of which theoretical model they follow. The range of values covered is pretty wide. All in all, the method extends the kinematic reach of usual searches, as it also does with the collection of final states.

An example distribution from MUSiC: Transverse mass for the final state comprising of one muon and missing transverse momentum. Color histograms: Simulated Standard Model processes. Red line: Signal from a hypothetical W’ boson with mass of 3TeV. (Source: paper)

So the kinematic distributions are checked against the simulated expectations in an automatized way, with MUSiC looking for every physicist’s dream: deviations. Any deviation from the simulation, meaning either fewer or more recorded events, is quantified by getting a probability value. This probability is calculated by also taking into account the much dreaded “look elsewhere effect”. (Which comes from the fact that, statistically, in a large number of distributions a random fluctuation that will mimic a genuine deviation is bound to appear sooner or later.)

When all’s said and done the collection of probabilities is overviewed. The MUSiC protocol says that any significant deviation will be scrutinized with more traditional methods – only that this need never actually arose in the 2016 data: all the data played along with the Standard Model, in all 1,069 examined final states and their kinematic ranges.

For the record, the largest deviation was spotted in the final state comprising three electrons, two generic hadronic jets and one jet coming from a bottom quark. Seven events were counted whereas the simulation gave 2.7±1.8 events (mostly coming from the production of a top plus an anti-top quark plus an intermediate vector boson from the collision; the fractional values are due to extrapolating to the amount of collected data). This excess was not seen in other related final states, “related” in that they also either include the same particles or have one less. Everything pointed to a fluctuation and the case was closed.

However, the goal of MUSiC was not strictly to find something new, but rather to demonstrate a method for model un-specific searches with collisions data. The mission seems to be accomplished, with CMS becoming even more general-purpose.

Read more:

Another generic search method in ATLAS: Going Rogue: The Search for Anything (and Everything) with ATLAS

And a take with machine learning: Letting the Machines Seach for New Physics

Fancy checking a good old model-specific search? Uncovering a Higgs Hiding Behind Backgrounds

A shortcut to truth

Article title: “Automated detector simulation and reconstruction
parametrization using machine learning”

Authors: D. Benjamin, S.V. Chekanov, W. Hopkins, Y. Li, J.R. Love

Reference: https://arxiv.org/abs/2002.11516 (https://iopscience.iop.org/article/10.1088/1748-0221/15/05/P05025)

Demonstration of probability density function as the output of a neural network. (Source: paper)

The simulation of particle collisions at the LHC is a pharaonic task. The messy chromodynamics of protons must be modeled; the statistics of the collision products must reflect the Standard Model; each particle has to travel through the detectors and interact with all the elements in its path. Its presence will eventually be reduced to electronic measurements, which, after all, is all we know about it.

The work of the simulation ends somewhere here, and that of the reconstruction starts; namely to go from electronic signals to particles. Reconstruction is a process common to simulation and to the real world. Starting from the tangle of statistical and detector effects that the actual measurements include, the goal is to divine the properties of the initial collision products.

Now, researchers at the Argonne National Laboratory looked into going from the simulated particles as produced in the collisions (aka “truth objects”) directly to the reconstructed ones (aka “reco objects”): bypassing the steps of the detailed interaction with the detectors and of the reconstruction algorithm could make the studies that use simulations much more speedy and efficient.

Display of a collision event involving hadronic jets at ATLAS. Each colored block corresponds to interaction with a detector element. (Source: ATLAS experiment)

The team used a neural network which it trained on simulations of the full set. The goal was to have the network learn to produce the properties of the reco objects when given only the truth objects. The process succeeded in producing the transverse momenta of hadronic jets, and looks suitable for any kind of particle and for other kinematic quantities.

More specifically, the researchers began with two million simulated jet events, fully passed through the ATLAS experiment and the reconstruction algorithm. For each of them, the network took the kinematic properties of the truth jet as input and was trained to achieve the reconstructed transverse momentum.

The network was taught to perform multi-categorization: its output didn’t consist of a single node giving the momentum value, but of 400 nodes, each corresponding to a different range of values. The output of each node was the probability for that particular range. In other words, the result was a probability density function for the reconstructed momentum of a given jet.

The final step was to select the momentum randomly from this distribution. For half a million of test jets, all this resulted in good agreement with the actual reconstructed momenta, specifically within 5% for values above 20 GeV. In addition, it seems that the training was sensitive to the effects of quantities other than the target one (e.g. the effects of the position in the detector), as the neural network was able to pick up on the dependencies between the input variables. Also, hadronic jets are complicated animals, so it is expected that the method will work on other objects just as well.

Comparison of the reconstructed transverse momentum between the full simulation and reconstruction (“Delphes”) and the neural net output. (Source: paper)

All in all, this work showed the perspective for neural networks to imitate successfully the effects of the detector and the reconstruction. Simulations in large experiments typically take up loads of time and resources due to their size, intricacy and frequent need for updates in the hardware conditions. Such a shortcut, needing only small numbers of fully processed events, would speed up studies such as optimization of the reconstruction and detector upgrades.

More reading:

Argonne Lab press release: https://www.anl.gov/article/learning-more-about-particle-collisions-with-machine-learning

Intro to neural networks: https://physicsworld.com/a/neural-networks-explained/

Letting the Machines Search for New Physics

Article: “Anomaly Detection for Resonant New Physics with Machine Learning”

Authors: Jack H. Collins, Kiel Howe, Benjamin Nachman

Reference : https://arxiv.org/abs/1805.02664

One of the main goals of LHC experiments is to look for signals of physics beyond the Standard Model; new particles that may explain some of the mysteries the Standard Model doesn’t answer. The typical way this works is that theorists come up with a new particle that would solve some mystery and they spell out how it interacts with the particles we already know about. Then experimentalists design a strategy of how to search for evidence of that particle in the mountains of data that the LHC produces. So far none of the searches performed in this way have seen any definitive evidence of new particles, leading experimentalists to rule out a lot of the parameter space of theorists favorite models.

A summary of searches the ATLAS collaboration has performed. The left columns show model being searched for, what experimental signature was looked at and how much data has been analyzed so far. The color bars show the regions that have been ruled out based on the null result of the search. As you can see, we have already covered a lot of territory.

Despite this extensive program of searches, one might wonder if we are still missing something. What if there was a new particle in the data, waiting to be discovered, but theorists haven’t thought of it yet so it hasn’t been looked for? This gives experimentalists a very interesting challenge, how do you look for something new, when you don’t know what you are looking for? One approach, which Particle Bites has talked about before, is to look at as many final states as possible and compare what you see in data to simulation and look for any large deviations. This is a good approach, but may be limited in its sensitivity to small signals. When a normal search for a specific model is performed one usually makes a series of selection requirements on the data, that are chosen to remove background events and keep signal events. Nowadays, these selection requirements are getting more complex, often using neural networks, a common type of machine learning model, trained to discriminate signal versus background. Without some sort of selection like this you may miss a smaller signal within the large amount of background events.

This new approach lets the neural network itself decide what signal to  look for. It uses part of the data itself to train a neural network to find a signal, and then uses the rest of the data to actually look for that signal. This lets you search for many different kinds of models at the same time!

If that sounds like magic, lets try to break it down. You have to assume something about the new particle you are looking for, and the technique here assumes it forms a resonant peak. This is a common assumption of searches. If a new particle were being produced in LHC collisions and then decaying, then you would get an excess of events where the invariant mass of its decay products have a particular value. So if you plotted the number of events in bins of invariant mass you would expect a new particle to show up as a nice peak on top of a relatively smooth background distribution. This is a very common search strategy, and often colloquially referred to as a ‘bump hunt’. This strategy was how the Higgs boson was discovered in 2012.

A histogram showing the invariant mass of photon pairs. The Higgs boson shows up as a bump at 125 GeV. Plot from here

The other secret ingredient we need is the idea of Classification Without Labels (abbreviated CWoLa, pronounced like koala). The way neural networks are usually trained in high energy physics is using fully labeled simulated examples. The network is shown a set of examples and then guesses which are signal and which are background. Using the true label of the event, the network is told which of the examples it got wrong, its parameters are updated accordingly, and it slowly improves. The crucial challenge when trying to train using real data is that we don’t know the true label of any of data, so its hard to tell the network how to improve. Rather than trying to use the true labels of any of the events, the CWoLA technique uses mixtures of events. Lets say you have 2 mixed samples of events, sample A and sample B, but you know that sample A has more signal events in it than sample B. Then, instead of trying to classify signal versus background directly, you can train a classifier to distinguish between events from sample A and events from sample B and what that network will learn to do is distinguish between signal and background. You can actually show that the optimal classifier for distinguishing the two mixed samples is the same as the optimal classifier of signal versus background. Even more amazing, this technique actually works quite well in practice, achieving good results even when there is only a few percent of signal in one of the samples.

An illustration of the CWoLa method. A classifier trained to distinguish between two mixed samples of signal and background events learns can learn to classify signal versus background. Taken from here

The technique described in the paper combines these two ideas in a clever way. Because we expect the new particle to show up in a narrow region of invariant mass, you can use some of your data to train a classifier to distinguish between events in a given slice of invariant mass from other events. If there is no signal with a mass in that region then the classifier should essentially learn nothing, but if there was a signal in that region that the classifier should learn to separate signal and background. Then one can apply that classifier to select events in the rest of your data (which hasn’t been used in the training) and look for a peak that would indicate a new particle. Because you don’t know ahead of time what mass any new particle should have, you scan over the whole range you have sufficient data for, looking for a new particle in each slice.

The specific case that they use to demonstrate the power of this technique is for new particles decaying to pairs of jets. On the surface, jets, the large sprays of particles produced when quark or gluon is made in a LHC collision, all look the same. But actually the insides of jets, their sub-structure, can contain very useful information about what kind of particle produced it. If a new particle that is produced decays into other particles, like top quarks, W bosons or some a new BSM particle, before decaying into quarks then there will be a lot of interesting sub-structure to the resulting jet, which can be used to distinguish it from regular jets. In this paper the neural network uses information about the sub-structure for both of the jets in event to determine if the event is signal-like or background-like.

The authors test out their new technique on a simulated dataset, containing some events where a new particle is produced and a large number of QCD background events. They train a neural network to distinguish events in a window of invariant mass of the jet pair from other events. With no selection applied there is no visible bump in the dijet invariant mass spectrum. With their technique they are able to train a classifier that can reject enough background such that a clear mass peak of the new particle shows up. This shows that you can find a new particle without relying on searching for a particular model, allowing you to be sensitive to particles overlooked by existing searches.

Demonstration of the bump hunt search. The shaded histogram is the amount of signal in the dataset. The different levels of blue points show the data remaining after applying tighter and tighter selection based on the neural network classifier score. The red line is the predicted amount of background events based on fitting the sideband regions. One can see that for the tightest selection (bottom set of points), the data forms a clear bump over the background estimate, indicating the presence of a new particle

This paper was one of the first to really demonstrate the power of machine-learning based searches. There is actually a competition being held to inspire researchers to try out other techniques on a mock dataset. So expect to see more new search strategies utilizing machine learning being released soon. Of course the real excitement will be when a search like this is applied to real data and we can see if machines can find new physics that us humans have overlooked!

Read More:

  1. Quanta Magazine Article “How Artificial Intelligence Can Supercharge the Search for New Particles”
  2. Blog Post on the CWoLa Method “Training Collider Classifiers on Real Data”
  3. Particle Bites Post “Going Rogue: The Search for Anything (and Everything) with ATLAS”
  4. Blog Post on applying ML to top quark decays “What does Bidirectional LSTM Neural Networks has to do with Top Quarks?”
  5. Extended Version of Original Paper “Extending the Bump Hunt with Machine Learning”

A new anomaly: the electromagnetic duality anomaly

Article: Electromagnetic duality anomaly in curved spacetimes
Authors: I. Agullo, A. del Rio and J. Navarro-Salas
Reference: arXiv:1607.08879

Disclaimer: this blogpost requires some basic knowledge of QFT (or being comfortable with taking my word at face value for some of the claims made :))

Anomalies exists everywhere. Probably the most intriguing ones are medical, but in particle physics they can be pretty fascinating too. In physics, anomalies refer to the breaking of a symmetry. There are basically two types of anomalies:

  • The first type, gauge anomalies, are red-flags: if they show up in your theory, they indicate that the theory is mathematically inconsistent.
  • The second type of anomaly does not signal any problems with the theory and in fact can have experimentally observable consequences. A prime example is the chiral anomaly. This anomaly nicely explains the decay rate of the neutral pion into two photons.
    Fig. 1: Illustration of pion decay into two photons. [Credit: Wikimedia Commons]

In this paper, a new anomaly is discussed. This anomaly is related to the polarization of light and is called the electromagnetic duality anomaly.

Chiral anomaly 101
So let’s first brush up on the basics of the chiral anomaly. How does this anomaly explain the decay rate of the neutral pion into two photons? For that we need to start with the Lagrangian for QED that describes the interactions between the electromagnetic field (that is, the photons) and spin-½ fermions (which pions are build from):

\displaystyle \mathcal L = \bar\psi \left( i \gamma^\mu \partial_\mu - i e \gamma^\mu A_\mu \right) \psi + m \bar\psi \psi

where the important players in the above equation are the \psis that describe the spin-½ particles and the vector potential A_\mu that describes the electromagnetic field. This Lagrangian is invariant under the chiral symmetry:

\displaystyle \psi \to e^{i \gamma_5} \psi .

Due to this symmetry the current density j^\mu = \bar{\psi} \gamma_5 \gamma^\mu \psi is conserved: \nabla_\mu j^\mu = 0. This then immediately tells us that the charge associated with this current density is time-independent. Since the chiral charge is time-independent, it prevents the \psi fields to decay into the electromagnetic fields, because the \psi field has a non-zero chiral charge and the photons have no chiral charge. Hence, if this was the end of the story, a pion would never be able to decay into two photons.

However, the conservation of the charge is only valid classically! As soon as you go from classical field theory to quantum field theory this is no longer true; hence, the name (quantum) anomaly.  This can be seen most succinctly using Fujikawa’s observation that even though the field \psi and Lagrangian are invariant under the chiral symmetry, this is not enough for the quantum theory to also be invariant. If we take the path integral approach to quantum field theory, it is not just the Lagrangian that needs to be invariant but the entire path integral needs to be:

\displaystyle \int D[A] \, D[\bar\psi]\, \int D[\psi] \, e^{i\int d^4x \mathcal L} .

From calculating how the chiral symmetry acts on the measure D \left[\psi \right]  \, D \left[\bar \psi \right], one can extract all the relevant physics such as the decay rate.

The electromagnetic duality anomaly
Just like the chiral anomaly, the electromagnetic duality anomaly also breaks a symmetry at the quantum level that exists classically. The symmetry that is broken in this case is – as you might have guessed from its name – the electromagnetic duality. This symmetry is a generalization of a symmetry you are already familiar with from source-free electromagnetism. If you write down source-free Maxwell equations, you can just swap the electric and magnetic field and the equations look the same (you just have to send  \displaystyle \vec{E} \to \vec{B} and \vec{B} \to - \vec{E}). Now the more general electromagnetic duality referred to here is slightly more difficult to visualize: it is a rotation in the space of the electromagnetic field tensor and its dual. However, its transformation is easy to write down mathematically:

\displaystyle F_{\mu \nu} \to \cos \theta \, F_{\mu \nu} + \sin \theta \, \, ^\ast F_{\mu \nu} .

In other words, since this is a symmetry, if you plug this transformation into the Lagrangian of electromagnetism, the Lagrangian will not change: it is invariant. Now following the same steps as for the chiral anomaly, we find that the associated current is conserved and its charge is time-independent due to the symmetry. Here, the charge is simply the difference between the number of photons with left helicity and those with right helicity.

Let us continue following the exact same steps as those for the chiral anomaly. The key is to first write electromagnetism in variables analogous to those of the chiral theory. Then you apply Fujikawa’s method and… *drum roll for the anomaly that is approaching*…. Anti-climax: nothing happens, everything seems to be fine. There are no anomalies, nothing!

So why the title of this blog? Well, as soon as you couple the electromagnetic field with a gravitational field, the electromagnetic duality is broken in a deeply quantum way. The number of photon with left helicity and right helicity is no longer conserved when your spacetime is curved.

Physical consequences
Some potentially really cool consequences have to do with the study of light passing by rotating stars, black holes or even rotating clusters. These astrophysical objects do not only gravitationally bend the light, but the optical helicity anomaly tells us that there might be a difference in polarization between lights rays coming from different sides of these objects. This may also have some consequences for the cosmic microwave background radiation, which is ‘picture’ of our universe when it was only 380,000 years old (as compared to the 13.8 billion years it is today!). How big this effect is and whether we will be able to see it in the near future is still an open question.

 

 

Further reading 

  • An introduction to anamolies using only quantum mechanics instead of quantum field theory is “Anomalies for pedestrians” by Barry Holstein 
  • The beautiful book “Quantum field theory and the Standard Model” by Michael Schwartz has a nice discussion in the later chapters on the chiral anomaly.
  • Lecture notes by Adal Bilal for graduate students on anomalies in general  can be found here

LIGO and Gravitational Waves: A Hep-ex perspective

The exciting Twitter rumors have been confirmed! On Thursday, LIGO finally announced the first direct observation of gravitational waves, a prediction 100 years in the making. The media storm has been insane, with physicists referring to the discovery as “more significant than the discovery of the Higgs boson… the biggest scientific breakthrough of the century.” Watching Thursday’s press conference from CERN, it was hard not to make comparisons between the discovery of the Higgs and LIGO’s announcement.

 

 

The gravitational-wave event GW150914 observed by the LIGO Collaboration
The gravitational-wave event GW150914 observed by the LIGO Collaboration

 

Long standing Searches for well known phenomena

 

The Higgs boson was billed as the last piece of the Standard Model puzzle. The existence of the Higgs was predicted in the 1960s in order to explain the mass of vector bosons of the Standard Model, and avoid non-unitary amplitudes in W boson scattering. Even if the Higgs didn’t exist, particle physicists expected new physics to come into play at the TeV Scale, and experiments at the LHC were designed to find it.

 

Similarly, gravitational waves were the last untested fundamental prediction of General Relativity. At first, physicists remained skeptical of the existence of gravitational waves, but the search began in earnest with Joseph Webber in the 1950s (Forbes). Indirect evidence of gravitational waves was demonstrated a few decades later. A binary system consisting of a pulsar and neutron star was observed to release energy over time, presumably in the form of gravitational waves. Using Webber’s method for inspiration, LIGO developed two detectors of unprecedented precision in order to finally make direct observation.

 

Unlike the Higgs, General Relativity makes clear predictions about the properties of gravitational waves. Waves should travel at the speed of light, have two polarizations, and interact weakly with matter. Scientists at LIGO were even searching for a very particular signal, described as a characteristic “chirp”. With the upgrade to the LIGO detectors, physicists were certain they’d be capable of observing gravitational waves. The only outstanding question was how often these observations would happen.

 

The search for the Higgs involved more uncertainties. The one parameter essential for describing the Higgs, its mass, is not predicted by the Standard Model. While previous collider experiments at LEP and Fermilab were able to set limits on the Higgs mass, the observed properties of the Higgs were ultimately unknown before the discovery. No one knew whether or not the Higgs would be a Standard Model Higgs, or part of a more complicated theory like Supersymmetry or technicolor.

 

Monumental scientific endeavors

 

Answering the most difficult questions posed by the universe isn’t easy, or cheap. In terms of cost, both LIGO and the LHC represent billion dollar investments. Including the most recent upgrade, LIGO cost a total $1.1 billion, and when it was originally approved in 1992, “it represented the biggest investment the NSF had ever made” according to France Córdova, NSF director. The discovery of the Higgs was estimated by Forbes to cost a total of $13 billion, a hefty price to be paid by CERN’s member and observer states. Even the electricity bill costs more than $200 million per year.

 

The large investment is necessitated by the sheer monstrosity of the experiments. LIGO consists of two identical detectors roughly 4 km long, built 3000 km apart. Because of it’s large size, LIGO is capable of measuring ripples in space 10000 times smaller than an atomic nucleus, the smallest scale ever measured by scientists (LIGO Fact Page). The size of the LIGO vacuum tubes is only surpassed by those at the LHC. At 27 km in circumference, the LHC is the single largest machine in the world, and the most powerful particle accelerator to date. It only took a handful of people to predict the existence of gravitational waves and the Higgs, but it took thousands of physicists and engineers to find them.

 

Life after Discovery

 

Even the language surrounding both announcements is strikingly similar. Rumors were circulating for months before the official press conferences, and the expectations from each respective community were very high. Both discoveries have been touted as the discoveries of the century, with many experts claiming that results would usher in a “new era” of particle physics or observational astronomy.

 

With a few years of hindsight, it is clear that the “new era” of particle physics has begun. Before Run I of the LHC, particle physicists knew they needed to search for the Higgs. Now that the Higgs has been discovered, there is much more uncertainty surrounding the field. The list of questions to try and answer is enormous. Physicists want to understand the source of the Dark Matter that makes up roughly 25% of the universe, from where neutrinos derive their mass, and how to quantize gravity. There are several ad hoc features of the Standard Model that merit additional explanation, and physicists are still searching for evidence of supersymmetry and grand unified theories. While the to-do list is long, and well understood, how to solve these problems is not. Measuring the properties of the Higgs does allow particle physicists to set limits on beyond the Standard Model Physics, but it’s unclear at which scale new physics will come into play, and there’s no real consensus about which experiments deserve the most support. For some in the field, this uncertainty can result in a great deal of anxiety and skepticism about the future. For others, the long to-do list is an absolutely thrilling call to action.

 

With regards to the LIGO experiment, the future is much more clear. LIGO has only published one event from 16 days of data taking. There is much more data already in the pipeline, and more interferometers like VIRGO and (e)LISA, planning to go online in the near future. Now that gravitational waves have been proven to exist, they can be used to observe the universe in a whole new way. The first event already contains an interesting surprise. LIGO has observed two inspriraling black holes of 36 and 29 solar masses, merging into a final black hole of 62 solar masses. The data thus confirmed the existence of heavy stellar black holes, with masses more than 25 times greater than the sun, and that binary black hole systems form in nature (Atrophysical Journal). When VIRGO comes online, it will be possible to triangulate the source of these gravitational waves as well. LIGO’s job is to watch, and see what other secrets the universe has in store.