Particle Physics Planet

October 31, 2014

arXiv blog

How Entanglement-Generating Satellites Will Make the Quantum Internet Global

Sending entangled photons to opposite sides of the planet will require a small fleet of orbiting satellites, say physicists.

October 31, 2014 12:57 AM

October 30, 2014

Christian P. Robert - xi'an's og

label switching in Bayesian mixture models

cover of Mixture Estimation and ApplicationsA referee of our paper on approximating evidence for mixture model with Jeong Eun Lee pointed out the recent paper by Carlos Rodríguez and Stephen Walker on label switching in Bayesian mixture models: deterministic relabelling strategies. Which appeared this year in JCGS and went beyond, below or above my radar.

Label switching is an issue with mixture estimation (and other latent variable models) because mixture models are ill-posed models where part of the parameter is not identifiable. Indeed, the density of a mixture being a sum of terms

\sum_{j=1}^k \omega_j f(y|\theta_i)

the parameter (vector) of the ω’s and of the θ’s is at best identifiable up to an arbitrary permutation of the components of the above sum. In other words, “component #1 of the mixture” is not a meaningful concept. And hence cannot be estimated.

This problem has been known for quite a while, much prior to EM and MCMC algorithms for mixtures, but it is only since mixtures have become truly estimable by Bayesian approaches that the debate has grown on this issue. In the very early days, Jean Diebolt and I proposed ordering the components in a unique way to give them a meaning. For instant, “component #1″ would then be the component with the smallest mean or the smallest weight and so on… Later, in one of my favourite X papers, with Gilles Celeux and Merrilee Hurn, we exposed the convergence issues related with the non-identifiability of mixture models, namely that the posterior distributions were almost always multimodal, with a multiple of k! symmetric modes in the case of exchangeable priors, and therefore that Markov chains would have trouble to visit all those modes in a symmetric manner, despite the symmetry being guaranteed from the shape of the posterior. And we conclude with the slightly provocative statement that hardly any Markov chain inferring about mixture models had ever converged! In parallel, time-wise, Matthew Stephens had completed a thesis at Oxford on the same topic and proposed solutions for relabelling MCMC simulations in order to identify a single mode and hence produce meaningful estimators. Giving another meaning to the notion of “component #1″.

And then the topic began to attract more and more researchers, being both simple to describe and frustrating in its lack of definitive answer, both from simulation and inference perspectives. Rodriguez’s and Walker’s paper provides a survey on the label switching strategies in the Bayesian processing of mixtures, but its innovative part is in deriving a relabelling strategy. Which consists of finding the optimal permutation (at each iteration of the Markov chain) by minimising a loss function inspired from k-means clustering. Which is connected with both Stephens’ and our [JASA, 2000] loss functions. The performances of this new version are shown to be roughly comparable with those of other relabelling strategies, in the case of Gaussian mixtures. (Making me wonder if the choice of the loss function is not favourable to Gaussian mixtures.) And somehow faster than Stephens’ Kullback-Leibler loss approach.

“Hence, in an MCMC algorithm, the indices of the parameters can permute multiple times between iterations. As a result, we cannot identify the hidden groups that make [all] ergodic averages to estimate characteristics of the components useless.”

One section of the paper puzzles me, albeit it does not impact the methodology and the conclusions. In Section 2.1 (p.27), the authors consider the quantity

p(z_i=j|{\mathbf y})

which is the marginal probability of allocating observation i to cluster or component j. Under an exchangeable prior, this quantity is uniformly equal to 1/k for all observations i and all components j, by virtue of the invariance under permutation of the indices… So at best this can serve as a control variate. Later in Section 2.2 (p.28), the above sentence does signal a problem with those averages but it seem to attribute it to MCMC behaviour rather than to the invariance of the posterior (or to the non-identifiability of the components per se). At last, the paper mentions that “given the allocations, the likelihood is invariant under permutations of the parameters and the allocations” (p.28), which is not correct, since eqn. (8)

f(y_i|\theta_{\sigma(z_i)}) =f(y_i|\theta_{\tau(z_i)})

does not hold when the two permutations σ and τ give different images of zi

Filed under: Books, Statistics, University life Tagged: component of a mixture, convergence, finite mixtures, identifiability, ill-posed problem, invariance, label switching, loss function, MCMC algorithms, missing data, multimodality, relabelling

by xi'an at October 30, 2014 11:14 PM

Emily Lakdawalla - The Planetary Society Blog

LightSail Vibration Test Shakes Loose New Problems
LightSail's random vibration test, meant to simulate the stress of an Atlas V rocket launch, shook loose new problems that the team will have to address.

October 30, 2014 07:56 PM

Clifford V. Johnson - Asymptotia

Loafing Around
Yes... loafing_around_1 I do still try to find time to make sure I slow down and make a batch of bread, roughly every week. The process of slowly kneading the dough, rolling and squeezing and folding again and again, is a good meditation. Then there's the reward of a house full of the smell of baking bread... [...] Click to continue reading this post

by Clifford at October 30, 2014 04:10 PM

Emily Lakdawalla - The Planetary Society Blog

Hayabusa 2 nearly ready for launch: Photos from Tanegashima, and new artist's renderings
On October 27, JAXA provided media with an opportunity to view the Hayabusa 2 spacecraft at the Tanegashima space center, where it's making final preparations for launch. Koumei Shibata was there, and took several photos. And artist Go Miyazaki has shared several terrific new renderings of the spacecraft in flight.

October 30, 2014 03:15 PM

astrobites - astro-ph reader's digest

In the beginning was Theia
Title: Dynamical Evolution of the Earth-Moon Progenitors – Whence Theia?
Authors: Billy L. Quarlesa & Jack J. Lissauera
First Author Institution: Space Science and Astrobiology Division MS 245-3, NASA Ames Research Center, Moffett Field, CA 94035, U.S.A.
Status: Accepted for publication in Icarus


Figure 1: Artist’s impression of the Giant Impact Hypothesis of the formation of the Moon.

The mystery of the formation of the Earth’s Moon was one that plagued astronomers for centuries. Many theories were proposed, including the fission of the Moon from the Earth’s crust, gravitational capture of the Moon and co-formation of the Earth and the Moon at the same time from the debris disk of material around the Sun. However neither of these theories successfully explained many of the oddities we observe in the Earth-Moon system; the lack of iron on the Moon compared to the iron rich Earth, the high-angular momentum of the system and the similar material composition of the two bodies, which are massively different from any other rocky Solar System body.

The Giant Impact Hypothesis says that the Moon was formed from the debris of a glancing collision between Earth and another rocky body roughly the size of Mars (1/2 the diameter of the Earth) named Theia (after the mythological Greek goddess who was the mother of the goddess of the Moon). It was proposed in the mid 20th century but was dismissed as an unlikely scenario until the 1980s when it emerged as the leading theory because it could better explain observations. The lack of iron in the Moon is caused by the impact, which immediately turns both colliding bodies molten – since iron is the heaviest element it sinks to the core of the Earth, leaving the Moon to form from the anaemic debris thrown off by both planets. This also explains why the Earth and Moon are so similar in composition, as the material from the Earth and Theia was so thoroughly mixed in the impact. The high angular momentum is also a relic of the large amount of energy input into the system by the impact.

As technology has progressed, simulations have shown that an impact at a glancing angle of 30 – 40° would result in an Earth-Moon system like we see today. But arriving at a scenario where Theia can impact upon the Earth at this very specific angle range in the turbulent early Solar System requires a very specific set of initial configurations of the Solar System. The authors of this paper try to determine what the possible configurations of orbital parameters of the 5 inner rocky planets (Mercury, Venus, Theia, Earth and Mars) could have been in order to provide the Goldilocks scenario for the formation of the Moon.

They put together a model of the Solar System from its late stages of formation describing each planet by its orbital parameters, such as the eccentricity (how elliptical the orbit is), semi-major axis (like a radius, the longest axis of the ellipse) and inclination (angle of the orbit with respect to the spin plane of the Sun) and also maker the following assumptions:

  1. That the sum of the masses of proto-Earth and Theia equals the sum of the current Earth-Moon system.
  2. That Theia originated from the general neighbourhood of the proto-Earth; it has a starting position somewhere between the orbit of Venus and just past the orbit of Mars.

Screenshot 2014-10-29 11.03.22

Figure 2: Schematic showing the semi-major axes of the 5 major planets in the inner Solar System (Mercury – orange, Venus – yellow, proto-Moon – blue, proto-Earth – green, Mars – red) which result in a “successful”

They begin their simulations in the late stages of formation of the Solar System (30-50 Myrs after the formation of the Sun) when most planetesimals (small, solid bodies in orbit around a star which will go on to form a planet) are expected to have been accreted onto one of the 5 larger planetary bodies. They therefore do not include any planetesimals  in their simulations. They investigate many different possible combinations of orbital parameters for each of the planets and observe which of these simulations result in an impact. They define a “successful” simulation when an impact occurs between 70-110 Myr after formation of the Sun to coincide with carbon dating of rocks on both the Earth and Moon.

Figure 3: Semi-major axis (a) against the eccentricity (e) of "successful" simulations producing an impact for the proto-Moon. Points are colour coded by their average AMD value which is an indicator of how much energy is left within the simulation system. This shows the variety of initial conditions that can produce an impact.

Figure 3: Semi-major axis (a) against the eccentricity (e) of “successful” simulations producing an impact for the proto-Moon. Points are colour coded by their average AMD value which is an indicator of how much energy is left within the simulation system. The different shaped icons specify which of the author’s simulations was used to produce the orbital parameters. This plot shows the variety of initial conditions that can produce an impact between the pro to-Earth and Theia.

The authors find that a significant fraction of the orbital parameters tested produce a “successful” outcome of a collision. The diagram in Figure 2 shows the semi-major axes of the “successful” scenarios where an impact occurs in the pre-defined time window.

The authors also consider the Angular Momentum Deficit (AMD) as a success measure of their simulations, which is an indicator of how much energy is left within the simulation system because it measures the system’s deviation from being circular and coplanar; this helps to identify which simulations are dynamically “cold” like the current day Solar System (AMD < 1.5). Figure 3 shows the eccentricity (e) and semi-major axis (a) of the proto-Moon colour coded by the mean AMD value of the simulation for all the runs producing a successful impact. This shows the variety of combinations of orbital parameters which can produce the right conditions for a Giant Impact to occur.

Perhaps then, such an impact would not have been such a rare occurrence after all. Perhaps Theia was just the unlucky one in our turbulent early Solar System, on a doomed collision course with Earth from the very beginning.



by Becky Smethurst at October 30, 2014 03:05 PM

Symmetrybreaking - Fermilab/SLAC

Reading the heavens with your phone

Two groups have released early versions of apps to turn your smart phone into a cosmic ray detector.

Cosmic rays, most of which come speeding through the Milky Way from outside of our solar system, crash into the Earth’s atmosphere at energies high enough to put the Large Hadron Collider to shame.

And all it takes to catch such an event is a smart phone.

Two groups are working on apps to turn smart phones into roving particle detectors. One group aims to educate, while the other is on a quest to create the largest cosmic ray detector array in the world.

Smart phone cameras contain sensors that help convert particles of light into the digital images that appear on your screen. Astronomers use high-powered versions of these sensors to study the light from faraway galaxies.

When a cosmic ray hits the Earth’s atmosphere, it produces a shower of energetic particles that rain down on the planet. When one of those particles hits a sensor, it leaves a temporary mark—usually a single hit pixel, but sometimes a multi-pixel streak.

If these events were more common or obtrusive, they’d be the scourge of the digital photography world. As it is, it takes a little more work to find them, something the app developers are happy to do.

Justin Vandenbroucke, a physicist at the University of Wisconsin, Madison, and Levi Simons, director of citizen science at the LA Makerspace in Los Angeles, lead one group, which for the past four years has been working to build an app that teachers and students can use to create their own cosmic ray experiments. It’s called DECO, Distributed Electronic Cosmic-ray Observatory.

“Instead of reading about particle interactions, students can see them on their own phones,” Vandenbroucke says. “We’re working with high school students to test the app, and we’re working with high school teachers to develop the curriculum.”

Vandenbroucke met Simons, then a physics-grad-student-turned-teacher, when they were paired together at SLAC National Accelerator Laboratory through STAR, a research experience program run through California Polytechnic State University. Their project began as an attempt to resurrect a defunct program in which students detected cosmic rays from the rooftops of schools in Southern California.

They say they envision students around the world comparing results. Young scientists in Madison and Boulder could compare data to examine the effect of altitude on cosmic ray detection. Vandenbroucke and Simons can also see students comparing their findings with magnetic field data to see if both are affected during a solar storm.

The other app-developing group has somewhat different plans.

Physicists Daniel Whiteson of the University of California, Irvine, and Michael Mulhearn of UC Davis met while working on competing experiments at Fermi National Accelerator Laboratory during grad school. They now work on competing experiments at the Large Hadron Collider at CERN.

For the past year, they have been designing an app called CRAYFIS that could turn a network of smart phones into the world’s largest cosmic ray shower detector. “Nobody’s built a device of this size before,” Whiteson says.

The two say that, if enough people in dense groups download the app, they will be able to detect multiple particles in showers caused by high-energy cosmic rays. Getting a clear picture could reveal information such as the original direction of the cosmic ray.

“Five hundred to 1000 phones per square kilometer is the density where we start being able to make measurements,” Mulhearn says.

The app can also run on tablets. Whiteson and Mulhearn posted a paper about the app to the arXiv last week.

Their app has the ability to run automatically when the light contamination is blocked and the phone is charging (to avoid draining the battery). The app sends data to a central server only when the phone is connected to wifi.

“The idea is to be very unobtrusive to the user,” Mulhearn says. “It will never complain. It just waits until the conditions are right.”

Vandenbroucke and Simons have received funding from the American Physical Society, the Knight Foundation and the Simon-Strauss Foundation. For Whiteson and Mulhearn, the work so far has been a labor of love.

The two groups, which have been working independently, recently made contact and have begun to discuss joining forces. 


Like what you see? Sign up for a free subscription to symmetry!

by Kathryn Jepsen at October 30, 2014 01:00 PM

The n-Category Cafe

Maths, Just in Short Words

Guest post by David Roberts

How much maths can you talk about if you could just use short words? Late one night, somewhere between waking and sleeping, the cartoon proof of Löb’s theorem and Boolos’ explanation of Gödel’s second incompleteness theorem (Paywall! But the relevant part is all on the first page, if you can see that) teamed up to induce me to produce a proof of Cantor’s theorem of the existence of more than one infinite cardinal, using only words of one syllable (or, “just short words”). I then had the idea that it would be interesting to collect explanations or definitions or proofs that used only words of one syllable, and perhaps publish them in one go. Below the fold, I give my proof of Cantor’s theorem, comments and criticisms welcomed. I was aiming for as complete a proof as I could, whereas a number of other people (more on that below) went for simplicity, or as short as they could, so the style is different to the others I mention.

The fact of Georg C., in short words.

  1. A “set” is made up of things.
  2. The set <semantics><annotation encoding="application/x-tex">\mathbb{N}</annotation></semantics> is made up of <semantics>1,2,3,4,,n,<annotation encoding="application/x-tex">1,2,3,4,\ldots,n,\ldots</annotation></semantics>
  3. We can add 1 to a thing in <semantics><annotation encoding="application/x-tex">\mathbb{N}</annotation></semantics> to get a new thing in <semantics><annotation encoding="application/x-tex">\mathbb{N}</annotation></semantics>, not the same as the first.
  4. A “map” from the set <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics> to the set <semantics>B<annotation encoding="application/x-tex">B</annotation></semantics> is a rule that takes a thing in <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics> and gives a thing in <semantics>B<annotation encoding="application/x-tex">B</annotation></semantics>.
  5. For a set <semantics>C<annotation encoding="application/x-tex">C</annotation></semantics> and a set <semantics>D<annotation encoding="application/x-tex">D</annotation></semantics>, the set <semantics>Map(C,D)<annotation encoding="application/x-tex">Map(C,D)</annotation></semantics>, “maps from <semantics>C<annotation encoding="application/x-tex">C</annotation></semantics> to <semantics>D<annotation encoding="application/x-tex">D</annotation></semantics>”, is made up of all maps from <semantics>C<annotation encoding="application/x-tex">C</annotation></semantics> to <semantics>D<annotation encoding="application/x-tex">D</annotation></semantics>.
  6. The set <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics> is said to be “not as big as” the set <semantics>B<annotation encoding="application/x-tex">B</annotation></semantics> if: when you give me a map from <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics> to <semantics>B<annotation encoding="application/x-tex">B</annotation></semantics>, I can find a thing in <semantics>B<annotation encoding="application/x-tex">B</annotation></semantics> that does not come, by the rule for the map, from a thing in <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics>.
  7. Let M be a map from the set <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics> to the set <semantics>Map(A,)<annotation encoding="application/x-tex">Map(A,\mathbb{N})</annotation></semantics>.
  8. This means that <semantics>M<annotation encoding="application/x-tex">M</annotation></semantics> is a rule that gives a map from <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics> to <semantics><annotation encoding="application/x-tex">\mathbb{N}</annotation></semantics>, for each thing <semantics>a<annotation encoding="application/x-tex">a</annotation></semantics> in <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics>. We will call this “<semantics>a<annotation encoding="application/x-tex">a</annotation></semantics>’s Map”, with the fact this is by the rule for <semantics>M<annotation encoding="application/x-tex">M</annotation></semantics> in the back of our minds.
  9. Now I will tell you a thing <semantics>C<annotation encoding="application/x-tex">C</annotation></semantics> in <semantics>Map(A,)<annotation encoding="application/x-tex">Map(A,\mathbb{N})</annotation></semantics>. It is a map from <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics> to <semantics><annotation encoding="application/x-tex">\mathbb{N}</annotation></semantics>, with rule:

    the thing <semantics>a<annotation encoding="application/x-tex">a</annotation></semantics> (in <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics>) goes to the thing that <semantics>a<annotation encoding="application/x-tex">a</annotation></semantics>’s Map sends <semantics>a<annotation encoding="application/x-tex">a</annotation></semantics> to (a thing in <semantics><annotation encoding="application/x-tex">\mathbb{N}</annotation></semantics>) plus 1.

  10. Read that once more, as it’s a bit hard to grasp.

  11. I claim that <semantics>C<annotation encoding="application/x-tex">C</annotation></semantics> is a thing that does not come from, by the rule for <semantics>M<annotation encoding="application/x-tex">M</annotation></semantics>, from a thing in <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics>.
  12. Why? Since if it did, let us say, come from the thing <semantics>a<annotation encoding="application/x-tex">a</annotation></semantics> in <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics>, then <semantics>C<annotation encoding="application/x-tex">C</annotation></semantics> is <semantics>a<annotation encoding="application/x-tex">a</annotation></semantics>’s Map for some <semantics>a<annotation encoding="application/x-tex">a</annotation></semantics>, and so the thing to which <semantics>a<annotation encoding="application/x-tex">a</annotation></semantics>’s Map sends <semantics>a<annotation encoding="application/x-tex">a</annotation></semantics> is the same thing to which <semantics>C<annotation encoding="application/x-tex">C</annotation></semantics> sends <semantics>a<annotation encoding="application/x-tex">a</annotation></semantics>, which is the thing to which <semantics>a<annotation encoding="application/x-tex">a</annotation></semantics>’s Map sends <semantics>a<annotation encoding="application/x-tex">a</annotation></semantics>, now plus 1, which is false, by point 3. So <semantics>C<annotation encoding="application/x-tex">C</annotation></semantics> is not <semantics>a<annotation encoding="application/x-tex">a</annotation></semantics>’s Map!
  13. So we were wrong to think that <semantics>C<annotation encoding="application/x-tex">C</annotation></semantics> was a thing that came from <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics> (by the rule for <semantics>M<annotation encoding="application/x-tex">M</annotation></semantics>), and so <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics> must be not as big as <semantics>Map(A,)<annotation encoding="application/x-tex">Map(A,\mathbb{N})</annotation></semantics>.
  14. If we take <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics> to be the set <semantics><annotation encoding="application/x-tex">\mathbb{N}</annotation></semantics>, this means that though <semantics><annotation encoding="application/x-tex">\mathbb{N}</annotation></semantics> is a big, big set, it is not as big as <semantics>Map(,)<annotation encoding="application/x-tex">Map(\mathbb{N},\mathbb{N})</annotation></semantics>!

This proof I learned on the homotopy type theory mailing list (this post by Thomas Streicher), and I like it as it uses (almost) minimal logical assumptions: no need for power sets, Axiom of Choice, excluded middle etc. There is of course Lawvere’s proof via his fixed-point theorem, which doesn’t even require function spaces, though it is less clear what this means for cardinalities. To me, though, it feels like an a proof in ‘external logic’, reasoning about the ambient category of discourse from the outside. The proof above works in the internal logic of any <semantics>Π<annotation encoding="application/x-tex">\Pi</annotation></semantics>-pretopos with natural number object, I believe. Probably it works if we replace the abstract object <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics> with <semantics><annotation encoding="application/x-tex">\mathbb{N}</annotation></semantics> and assume that <semantics> <annotation encoding="application/x-tex">\mathbb{N}^\mathbb{N}</annotation></semantics> exists, rather than assuming cartesian closedness. But this is off-topic.

After I mentioned this on Google+, Joel Hamkins sent me a different proof of Cantor’s theorem, all in single-syllable words, then a proof of the irrationality of <semantics>2<annotation encoding="application/x-tex">\sqrt{2}</annotation></semantics>. Then Tim Gowers wrote another irrationality proof, and then, on my prompting, an explanation and outline of the proof of the Green-Tao theorem on arithmetic progressions in the primes. Toby Bartels wrote out the Peano axioms. Asaf Karagila wrote a proof of Cantor’s theorem without using the letter ‘e’. You can can a number of these in the comments at my original Google+ post, but read on before clicking through.

What I’d really like to see is two things. First: people who have proved amazing theorems explain said theorems in this manner (I was hoping to get, for instance, Terry Tao or Ben Green to write up the Green-Tao theorem). Or perhaps see how many Fields Medallists we could get to do this exercise, on the result/concept of their choice. Second, and this is a more local challenge to Café patrons: write about some category-theoretic concept using monosyllabic words. Otherwise, anyone can chip in with their favourite proof, fact, concept or definition in the comments! It is best to give it a go before reading others’ attempts, especially if you happen to have chosen the same thing to write about as someone else.

So get those pens or keys and go: write maths in just short words!

by david ( at October 30, 2014 11:50 AM

Lubos Motl - string vacua and pheno

NPR releases another anti-SUSY rant
Exactly half a year ago, Joe Lykken and Maria Spiropulu printed an unwise anti-physics diatribe in Scientific American. In order to prove that it's not weaker in similar fashionable attacks, NPR published its own rant written by a professional critic of science named Marcelo Gleiser:
Are Physicists Ready To Give Up The Chase For SUSY?
It was at the end of April 2014. Now it's the end of October 2014 and NPR printed an anti-SUSY article written by a man named Marcelo Gleiser:
Can Scientific Belief Go Too Far?
If your child couldn't understand what "deja vu" means, maybe you could use these two articles as an example. You may also use these articles to explain the slogan by Joseph Goebbels, "a lie repeated 100 times becomes the truth".

In the new rant, Gleiser suggests that the "belief in science" or "particular scientific beliefs" are analogous to the "religious beliefs". Well, yes or no: the devil is in the details. And most of the details that Gleiser tries to sell are profoundly misleading or downright wrong. He doesn't really understand (or at least neglects, when it comes to particular questions) most of the features by which science differs from religion – and in fact, he tries to force science to accept his quasi-religious ideas how people should decide what the truth is.

To mention something that is right about the article, Gleiser seems aware – at least when it comes to his general words – of the difference that science doesn't allow "absolute dogmas". If something is demonstrably wrong, the amount of evidence ultimately becomes overwhelming enough for every scientist to see that it was wrong. But this "detail" makes quite a difference.

However, everything else that Gleiser writes seems to unmask his misunderstanding what science is and what it is not – and especially what science has to obey and what it doesn't have to obey.
Of course, the product of scientific research must be something concrete: Hypotheses must be either confirmed or refuted, and data from experiments should be repeatable by others. Penicillin does cure diseases, airplanes fly and Halley's comet does come back every 76 years.
The important breakthroughs in modern physics just didn't follow and couldn't have followed this template. A problem with the naive caricature of science above is that in the actual scientific process, the construction of a "viable guess" is actually the most nontrivial part of the whole scientific process, the work for which the theorists get most of their salaries, and (in the happy cases) the achievements for which the most famous scientists are celebrated – while the quote above suggests that it is some triviality, some cheap guesswork at the beginning, and that the "bulk" of the scientific process only occurs later when the guess is being verified.

The process of the "discovery of the special theory of relativity" wasn't really about easily guessing the final form of the theory, and then working hard for many years to verify it. By the time when the final form of special relativity was completed – when the "right guess" or the "viable hypothesis" became available to Albert Einstein – the discovery had already been pretty much completed! The point is that the hypotheses are being constantly modified and have to be constantly modified as new experiments as well as theoretical realizations are being taken into account. And a genius is often needed to make the really big leaps.

My example focused on special relativity; the development of general relativity or quantum mechanics would be even better examples to prove my point. But a similar comments holds for pretty much every important and conceptually deep breakthrough in physics.

Gleiser – and many other people – just don't understand a key fact of the scientific method, namely the fact that the knowledge is constantly evolving which is why the hypotheses that are being developed are evolving, too. It has to be so. He presents this very fact as an "illness" of science when it is done incorrectly. A "good science", as he envisions it, only works with fixed hypotheses. But the truth is exactly the opposite. Science has to work with constantly evolving hypotheses and this is actually a key feature that makes science superior in comparison with religion!

Gleiser also says:
This kind of posture, when there is a persistent holding on to a belief that is continually contradicted by facts, can only be called faith.
His example is the belief in realism – the anti-quantum zeal – held by Albert Einstein, Max Planck, Erwin Schrödinger, and others. Well, Einstein and pals were simply wrong about a scientific issue. I find it deeply misleading and counterproductive when well-defined scientific propositions are being sold as religion.

Einstein et al. could have had "quasi-religious" reasons not to accept the new quantum, probabilistic, non-realist framework of modern physics. But whether one calls such attitudes "religious" is a matter of terminology. What's important is
  1. that Einstein and pals had some rationally, scientifically justifiable but vague reasons for their intellectual inertia;
  2. that Einstein and pals were demonstrably wrong.
The fact that Einstein was imagining that classical physics had to underlie the quantum phenomena was analogous to religion only to the extent that every belief may be called "faith" and compared to religion. However, the reasons why Einstein believed such things were very different from the reasons why people believe in Jesus Christ. People believe in Jesus Christ because they have read texts or heard about this man from others; on the other hand, Einstein had beliefs about the hypothetical foundations of the laws of physics because he extrapolated lots of previous completely scientific successes of completely scientific discoveries confirming completely scientific theories.

It just happens that such extrapolations are unreliable and in this case and many others, they turned out and often turn out to be wrong. Einstein didn't actually have any "solid evidence" that classical deterministic physics was destined to survive. And indeed, we know that it couldn't have survived. On the other hand, the observation that a certain form of the laws of physics has been successful for centuries to explain thousands of phenomena isn't a religious observation. It is an observation of a scientific kind, a piece of circumstantial evidence that legitimately affects a scientist's belief (and that may mislead him if the belief is wrong – which is often so if the scientist is overlooking much stronger evidence pointing to the opposite direction).

A more general type of this reasoning – some extrapolation of the lessons we have learned in the past – is absolutely critical for science. The whole history of science may be presented as an ever more accurate interpolation and extrapolation of the previously found laws. When Galileo measured the distance \(s(t)\) traveled by a freely falling object after time \(t\), he measured \(s(t)\) for integer values of \(t\) only (in some units, in Pisa). He made a guess about the functional dependence,\[

s(t) = \frac{gt^2}{2},

\] which already involves some interpolation – a form of generalization of some formulae to values of \(t\) for which no measurement has been made. Interpolation of functions is a very trivial example. Things become "just a bit more complex" when the functions depend on many variables, perhaps including some integer-valued and "more qualitative" ones. But science in general and physics in particular has been extending the previous insights to situations that may be called "qualitatively different". As it became increasingly capable of describing diverse phenomena with the same underlying laws, it was degrading situations that used to look as "qualitatively different ones" to situations that are "mutually analogous manifestations of the same laws".

If Gleiser wants to ban this extrapolation of the previous lessons of physics because it's "faith", he is surely throwing the baby out with the bath water.

Science simply couldn't work without this kind of interpolation and extrapolation. Interpolation and extrapolation – done in ever more abstract, far-reaching ways – is the most important method in all of science to produce the hypotheses that may be tested, confirmed, or serve as a basis for refinements when they fail the tests. When I read Gleiser's text, it seems pretty clear that he is not getting any of these things. For him, any extrapolation of the previous lessons we learned from science makes science as unjustified as religion. Again, the truth is the other way around: the scientific habit of making guesses about new and unknown questions that are rooted in our previous experience is a specific virtue of science that religions don't share. To sell this virtue as a vice is a complete misunderstanding of the scientific method.

But the most atrocious part of Gleiser's rant begin when he reiterates some common laymen's misconceptions about the "testability":
Bringing things to the present, we are currently going through a curious moment in high-energy physics, where some very popular theories may not be testable. This means that we can't determine whether they are wrong, which flies in the face of what science is about.
Similar aßholes got so used to writing similar trash that it has turned into a dogma and they became literally incapable of seeing how completely wrong these comments are. The truth is that if we can't determine whether a proposition is wrong, it surely doesn't mean that the proposition is untestable. For a proposition to be untestable, it would have to be impossible for any agent that may exist in the Universe now or in the future to test this proposition.

An overwhelming majority of the humans can't test the claim that the Moon has a surface on the other side that we don't see; that a muon decays to an electron and neutrinos; and virtually everything else. Even collectively, we can't really observe the core of the Sun and do millions of other things. But that doesn't mean that propositions about these objects and processes are untestable. It only means that they are untestable with certain very specific and very limited tools which is an entirely different characteristic, one that is surely not needed and was never needed for a question to be investigated by legitimate scientists.

Our ability to connect the observations with the scientific theories is improving in the long run. Most things that particle physicists are seeing today would look like untestable propositions by existing tools just a century (and sometimes decades) ago. Obviously, this fact couldn't have meant that there was something wrong about scientists who studied those things.
Like a zombie that never dies, it's possible to come up with theories that can always be redefined to escape the reach of current experiments. Case in point: supersymmetry, a hypothetical theory where each particle of matter (electrons, quarks) gains a supersymmetric partner.
Holy cow. Give me a Lagrangian of a quantum field theory and I will tell you whether it is supersymmetric or not. This decision is much sharper and much more well-defined and more unequivocal than anything that a sleazy jellyfish such as Marcelo Gleiser has ever investigated – or will ever investigate – in his gelatinous life.

Once again, the comment about "redefining" shows that Gleiser wants science to work like religions, focusing on the original dogmas all the time. I see no other way to interpret his visibly negative labels (zombies, escaping etc.) that he associates with "redefining". In proper science, however, this "redefinition" is a vital and welcome process that pushes the theories closer to the truth so it is absolutely idiotic to criticize science for this process.

Some regions of the supersymmetry parameter spaces have been ruled out, some remaining points of the supersymmetry parameter space are more accessible to the experiments in the near future, while some of them are less accessible. But this characteristic isn't an intrinsic quality of the theory itself. It is a quality that depends both on the theory and the current state of the technological progress. And this "testability by contemporary experiment" is something completely different than the question whether the theory – or the point of its parameter space – is natural or fine-tuned.

Whether a collection of values of parameters in a theory is natural, and therefore likely according to the naturalness paradigm, may be decided by looking at the theory and the numbers themselves. One doesn't have to think about the contemporary experiments at all! And be sure, lots of very natural scenarios that do assume supersymmetry are still alive and kicking, compatible with all the experiments we have. There is nothing zombie-like about these supersymmetric scenarios whatsoever.

Instead, what Mr Gleiser and other demagogues want to do is to "verbally connect" supersymmetry with the "new blasphemies" and the non-supersymmetric models with the "old good dogmas", and prefer the latter. But this dogmatic, asymmetric treatment of two hypotheses is exactly what is not allowed in science. Both supersymmetric models (some of the viable scenarios) and non-supersymmetric models (effective quantum field theories) exist that are compatible with all the observations we have made so far – so they must be considered as possibilities. It doesn't matter at all which of the competing explanations was proposed earlier and which of them is newer. What matters is whether they are compatible with the evidence. The relative likelihood of both scenarios depends on many things – and if there is an asymmetry here, one must realize that the smarter a person is, the more he thinks that the supersymmetric models have a higher prior probability than the non-supersymmetric ones.

Be sure that someone who is willing to misinterpret the current situation as something that eliminates SUSY even though it demonstrably doesn't is dishonest or an imbecile. Sadly, Mr Gleiser is both.

by Luboš Motl ( at October 30, 2014 10:27 AM

Peter Coles - In the Dark

Ice Watch

I thought I’d share this video about an installation called Ice Watch, which involves one hundred tonnes of inland ice from Greenland meltinging on the Radhusplads, Copenhagen’s City Hall Square. With Ice Watch, Olafur Eliasson and Minik Rosing direct attention to the publication of the IPCC’s 5th Assessment Report on the Earth’s Climate. The ice now melted, which happened faster than expected owing to the unusually warm weather for this time of year…




by telescoper at October 30, 2014 08:56 AM

Christian P. Robert - xi'an's og

Relevant statistics for Bayesian model choice [hot off the press!]

jrssbabcOur paper about evaluating statistics used for ABC model choice has just appeared in Series B! It somewhat paradoxical that it comes out just a few days after we submitted our paper on using random forests for Bayesian model choice, thus bypassing the need for selecting those summary statistics by incorporating all statistics available and letting the trees automatically rank those statistics in term of their discriminating power. Nonetheless, this paper remains an exciting piece of work (!) as it addresses the more general and pressing question of the validity of running a Bayesian analysis with only part of the information contained in the data. Quite usefull in my (biased) opinion when considering the emergence of approximate inference already discussed on this ‘Og…

[As a trivial aside, I had first used fresh from the press(es) as the bracketted comment, before I realised the meaning was not necessarily the same in English and in French.]

Filed under: Books, Statistics, University life Tagged: ABC model choice, Approximate Bayesian computation, JRSSB, Royal Statistical Society, Series B, statistical methodology, summary statistics

by xi'an at October 30, 2014 08:09 AM

Emily Lakdawalla - The Planetary Society Blog

The Antares Accident: Whose Rocket Was It?
Despite some in the media declaring it a NASA rocket disaster, Antares represents a new way of doing business. It's owned by a private company providing a service to NASA to resupply the space station. How is this different from other rockets NASA uses?

October 30, 2014 06:21 AM

John Baez - Azimuth

Sensing and Acting Under Information Constraints

I’m having a great time at a workshop on Biological and Bio-Inspired Information Theory in Banff, Canada. You can see videos of the talks online. There have been lots of good talks so far, but this one really blew my mind:

• Naftali Tishby, Sensing and acting under information constraints—a principled approach to biology and intelligence, 28 October 2014.

Tishby’s talk wasn’t easy for me to follow—he assumed you already knew rate-distortion theory and the Bellman equation, and I didn’t—but it was great!

It was about the ‘action-perception loop':

This is the feedback loop in which living organisms—like us—take actions depending on our goals and what we perceive, and perceive things depending on the actions we take and the state of the world.

How do we do this so well? Among other things, we need to balance the cost of storing information about the past against the payoff of achieving our desired goals in the future.

Tishby presented a detailed yet highly general mathematical model of this! And he ended by testing the model on experiments with cats listening to music and rats swimming to land.

It’s beautiful stuff. I want to learn it. I hope to blog about it as I understand more. But for now, let me just dive in and say some basic stuff. I’ll start with the two buzzwords I dropped on you. I hate it when people use terminology without ever explaining it.

Rate-distortion theory

Rate-distortion theory is a branch of information theory which seeks to find the minimum rate at which bits must be communicated over a noisy channel so that the signal can be approximately reconstructed at the other end without exceeding a given distortion. Shannon’s first big result in this theory, the ‘rate-distortion theorem’, gives a formula for this minimum rate. Needless to say, it still requires a lot of extra work to determine and achieve this minimum rate in practice.

For the basic definitions and a statement of the theorem, try this:

• Natasha Devroye, Rate-distortion theory, course notes, University of Chicago, Illinois, Fall 2009.

One of the people organizing this conference is a big expert on rate-distortion theory, and he wrote a book about it.

• Toby Berger, Rate Distortion Theory: A Mathematical Basis for Data Compression, Prentice–Hall, 1971.

Unfortunately it’s out of print and selling for $259 used on Amazon! An easier option might be this:

• Thomas M. Cover and Joy A. Thomas, Elements of Information Theory, Chapter 10: Rate Distortion Theory, Wiley, New York, 2006.

The Bellman equation

The Bellman equation reduces the task of finding an optimal course of action to choosing what to do at each step. So, you’re trying to maximize the ‘total reward’—the sum of rewards at each time step—and Bellman’s equation says what to do at each time step.

If you’ve studied physics, this should remind you of how starting from the principle of least action, we can get a differential equation describing the motion of a particle: the Euler–Lagrange equation.

And in fact they’re deeply related. The relation is obscured by two little things. First, Bellman’s equation is usually formulated in a context where time passes in discrete steps, while the Euler–Lagrange equation is usually formulated in continuous time. Second, Bellman’s equation is really the discrete-time version not of the Euler–Lagrange equation but a more or less equivalent thing: the ‘Hamilton–Jacobi equation’.

Ah, another buzzword to demystify! I was scared of the Hamilton–Jacobi equation for years, until I taught a course on classical mechanics that covered it. Now I think it’s the greatest thing in the world!

Briefly: the Hamilton–Jacobi equation concerns the least possible action to get from a fixed starting point to a point q in space at time t. If we call this least possible action W(t,q), the Hamilton–Jacobi equation says

\displaystyle{ \frac{\partial W(t,q)}{\partial q_i} = p_i  }

\displaystyle{ \frac{\partial W(t,q)}{\partial t} = -E  }

where p is the particle’s momentum at the endpoint of its path, and E is its energy there.

If we replace derivatives by differences, and talk about maximizing total reward instead of minimizing action, we get Bellman’s equation:

Bellman equation, Wikipedia.

Markov decision processes

Bellman’s equation can be useful whenever you’re trying to figure out an optimal course of action. An important example is a ‘Markov decision process’. To prepare you for Tishby’s talk, I should say what this is.

In a Markov process, something randomly hops around from state to state with fixed probabilities. In the simplest case there’s a finite set S of states, and time proceeds in discrete steps. At each time step, the probability to hop from state s to state s' is some fixed number P(s,s').

This sort of thing is called a Markov chain, or if you feel the need to be more insistent, a discrete-time Markov chain.

A Markov decision process is a generalization where an outside agent gets to change these probabilities! The agent gets to choose actions from some set A. If at a given time he chooses the action \alpha \in A, the probability of the system hopping from state s to state s' is P_\alpha(s,s'). Needless to say, these probabilities have to sum to one for any fixed s.

That would already be interesting, but the real fun is that there’s also a reward R_\alpha(s,s'). This is a real number saying how much joy or misery the agent experiences if he does action \alpha and the system hops from s to s'.

The problem is to choose a policy—a function from states to actions—that maximizes the total expected reward over some period of time. This is precisely the kind of thing Bellman’s equation is good for!

If you’re an economist you might also want to ‘discount’ future rewards, saying that a reward n time steps in the future gets multiplied by \gamma^n, where 0 < \gamma \le 1 is some discount factor. This extra tweak is easily handled, and you can see it all here:

Markov decision process, Wikipedia.

Partially observable Markov decision processes

There’s a further generalization where the agent can’t see all the details of the system! Instead, when he takes an action \alpha \in A and the system hops from state s to state s', he sees something: a point in some set O of observations. He makes the observation o \in O with probability \Omega_\alpha(o,s').

(I don’t know why this probability depends on s' but not s. Maybe it ultimately doesn’t matter much.)

Again, the goal is to choose a policy that maximizes the expected total reward. But a policy is a bit different now. The action at any time can only depend on all the observations made thus far.

Partially observable Markov decision processes are also called POMPDs. If you want to learn about them, try these:

Partially observable Markov decision process, Wikipedia.

• Tony Cassandra, Partially observable Markov decision processes.

The latter includes an introduction without any formulas to POMDPs and how to choose optimal policies. I’m not sure who would study this subject and not want to see formulas, but it’s certainly a good exercise to explain things using just words—and it makes certain things easier to understand (though not others, in a way that depends on who is trying to learn the stuff).

The action-perception loop

I already explained the action-perception loop, with the help of this picture from the University of Bielefeld’s Department of Cognitive Neuroscience:

Nafthali Tishby has a nice picture of it that’s more abstract:

We’re assuming time comes in discrete steps, just to keep things simple.

At each time t

• the world has some state W_t, and
• the agent has some state M_t.

Why the letter M? This stands for memory: it can be the state of the agent’s memory, but I prefer to think of it as the state of the agent.

At each time

• the agent takes an action A_t, which affects the world’s next state, and

• the world provides a sensation S_t to the agent, which affect’s the agent’s next state.

This is simplified but very nice. Note that there’s a symmetry interchanging the world and the agent!

We could make it fancier by having lots of agents who all interact, but there are a lot of questions already. The big question Tishby focuses on is optimizing how much the agent should remember about the past if they

• get a reward depending on the action taken and the resulting state of the world


• pay a price for the information stored from sensations.

Tishby formulates this optimization question as something like a partially observed Markov decision process, uses rate-distortion theory to analyze how much information needs to be stored to achieve a given reward, and uses Bellman’s equation to solve the optimization problem!

So, everything I sketched fits together somehow!

I hope what I’m saying now is roughly right: it will take me more time to get the details straight. If you’re having trouble absorbing all the information I just threw at you, don’t feel bad: so am I. But the math feels really natural and good to me. It involves a lot of my favorite ideas (like generalizations of the principle of least action, and relative entropy), and it seems ripe to be combined with network theory ideas.

For details, I highly recommend this paper:

• Naftali Tishby and Daniel Polani, Information theory of decisions and actions, in Perception-Reason-Action Cycle: Models, Algorithms and System. Vassilis, Hussain and Taylor, Springer, Berlin, 2010.

I’m going to print this out, put it by my bed, and read it every night until I’ve absorbed it.

by John Baez at October 30, 2014 12:04 AM

October 29, 2014

Quantum Diaries

Costumes to make zombie Einstein proud

This article appeared in symmetry on Oct. 21, 2014.

These physics-themed Halloween costume ideas are sure to entertain—and maybe even educate. Terrifying, we know. Image: Sandbox Studio, Chicago with Corinne Mucha

These physics-themed Halloween costume ideas are sure to entertain—and maybe even educate. Terrifying, we know. Image: Sandbox Studio, Chicago with Corinne Mucha


So you haven’t picked a Halloween costume, and the big night is fast approaching. If you’re looking for something a little funny, a little nerdy and sure to impress fellow physics fans, look no further. We’ve got you covered.

1. Dark energy

This is an active costume, perfect for the party-goer who plans to consume a large quantity of sugar. Suit up in all black or camouflage, then spend your evening squeezing between people and pushing them apart.

Congratulations! You’re dark energy: a mysterious force causing the accelerating expansion of the universe, intriguing in the lab and perplexing on the dance floor.

2. Cosmic inflation

Theory says that a fraction of a second after the big bang, the universe grew exponentially, expanding so that tiny fluctuations were stretched into the seeds of entire galaxies.

But good luck getting that costume through the door.

Instead, take a simple yellow life vest and draw the cosmos on it: stars, planets, asteroids, whatever you fancy. When friends pull on the emergency tab, the universe will grow.

3. Heisenberg Uncertainty Principle

Here’s a great excuse to repurpose your topical Breaking Bad costume from last year.

Walter White—aka “Heisenberg”—may have been a chemistry teacher, but the Heisenberg Uncertainty Principle is straight out of physics. Named after Werner Heisenberg, a German physicist credited with the creation of quantum mechanics, the Heisenberg Uncertainty Principle states that the more accurately you know the position of a particle, the less information you know about its momentum.

Put on Walter White’s signature hat and shades (or his yellow suit and respirator), but then add some uncertainty by pasting Riddler-esque question marks to your outfit.

4. Bad neutrino

A warning upfront: Only the ambitious and downright extroverted should attempt this costume.

Neutrinos are ghostly particles that pass through most matter undetected. In fact, trillions of neutrinos pass through your body every second without your knowledge.

But you aren’t going to go as any old neutrino. Oh no. You’re a bad neutrino—possibly the worst one in the universe—so you run into everything: lampposts, trees, haunted houses and yes, people. Don a simple white sheet and spend the evening interacting with everyone and everything.

5. Your favorite physics experiment

You physics junkies know that there are a lot of experiments with odd acronyms and names that are ripe for Halloween costumes. You can go as ATLAS (experiment at the Large Hadron Collider / character from Greek mythology), DarkSide (dark matter experiment at Gran Sasso National Laboratory / good reason to repurpose your Darth Vader costume), PICASSO (dark matter experiment at SNOLAB / creator of Cubism), MINERvA (Fermilab neutrino experiment / Roman goddess of wisdom), or the Dark Energy Survey (dark energy camera located at the Blanco Telescope in Chile / good opportunity for a pun).

Physics-loving parents can go as explorer Daniel Boone, while the kids go as neutrino experiments MicroBooNE and MiniBooNE. The kids can wear mini fur hats of their own or dress as detector tanks to be filled with candy.

6. Feynman diagram

You might know that a Feynman diagram is a drawing that uses lines and squiggles to represent a particle interaction. But have you ever noticed that they sometimes look like people? Try out this new take on the black outfit/white paint skeleton costume. Bonus points for going as a penguin diagram.

7. Antimatter

Break out the bell-bottoms and poster board. In bold letters, scrawl the words of your choosing: “I hate things!,” “Stuff is awful!,” and “Down with quarks!” will all do nicely. Protest from house to house and declare with pride that you are antimatter. It’s a fair critique: Physicists still aren’t sure why matter dominates the universe when equal amounts of matter and antimatter should have been created in the big bang.

Fortunately, you don’t have to solve this particular puzzle on your quest for candy. Just don’t high five anyone; you might annihilate.

8. Entangled particles

Einstein described quantum entanglement as “spooky action at a distance”—the perfect costume for Halloween. Entangled particles are extremely strange. Measuring one automatically determines the state of the other, instantaneously.

Find someone you are extremely in tune with and dress in opposite colors, like black and white. When no one is observing you, you can relax. But when interacting with people, be sure to coordinate movements. They spin to the left, you spin to the right. They wave with the right hand? You wave with the left. You get the drill.

You can also just wrap yourselves together in a net. No one said quantum entanglement has to be hard.

9. Holographic you(niverse)

The universe may be like a hologram, according to a theory currently being tested at Fermilab’s Holometer experiment. If so, information about spacetime is chunked into 2-D bits that only appear three-dimensional from our perspective.

Help others imagine this bizarre concept by printing out a photo of yourself and taping it to your front. You’ll still technically be 3-D, but that two-dimensional picture of your face will still start some interesting discussions. Perhaps best not to wear this if you have a busy schedule or no desire to discuss the nature of time and space while eating a Snickers.

10. Your favorite particle

There are many ways to dress up as a fundamental particle. Bring a lamp along to trick-or-treat to go as the photon, carrier of light. Hand out cookies to go as the Higgs boson, giver of mass. Spend the evening attaching things to people to go as a gluon.

To branch out beyond the Standard Model of particle physics, go as a supersymmetric particle, or sparticle: Wear a gladiator costume and shout, “I am Sparticle!” whenever someone asks about your costume.

Or grab a partner to become a meson, a particle made of a quark and antiquark. Mesons are typically unstable, so whenever you unlink arms, be sure to decay in a shower of electrons and neutrinos—or candy corn.

Lauren Biron

by Fermilab at October 29, 2014 04:46 PM

Sean Carroll - Preposterous Universe

The Science of Interstellar

The intersection — maybe the union! — of science and sci-fi geekdom is overcome with excitement about the upcoming movie Interstellar, which opens November 7. It’s a collaboration between director Christopher Nolan and physicist Kip Thorne, both heroes within their respective communities. I haven’t seen it yet myself, nor do I know any secret scoop, but there’s good reason to believe that this film will have some of the most realistic physics of any recent blockbuster we’ve seen. If it’s a success, perhaps other filmmakers will take the hint?

Kip, who is my colleague at Caltech (and a former guest-blogger), got into the science-fiction game quite a while back. He helped Carl Sagan with some science advice for his book Contact, later turned into a movie starring Jodie Foster. In particular, Sagan wanted to have some way for his characters to traverse great distances at speeds faster than light, by taking a shortcut through spacetime. Kip recognized that a wormhole was what was called for, but also realized that any form of faster-than-light travel had the possibility of leading to travel backwards in time. Thus was the entire field of wormhole time travel born.

As good as the movie version of Contact was, it still strayed from Sagan’s original vision, as his own complaints show. (“Ellie disgracefully waffles in the face of lightweight theological objections to rationalism…”) Making a big-budget Hollywood film is necessarily a highly collaborative endeavor, and generally turns into a long series of forced compromises. Kip has long been friends with Lynda Obst, an executive producer on Contact, and for years they batted around ideas for a movie that would really get the science right.

Long story short, Lynda and Kip teamed with screenwriter Jonathan Nolan (brother of Christopher), who wrote a draft of a screenplay, and Christopher eventually agreed to direct. I know that Kip has been very closely involved with the script as the film has developed, and he’s done his darnedest to make sure the science is right, or at least plausible. (We don’t actually whether wormholes are allowed by the laws of physics, but we don’t know that they’re not allowed.) But it’s a long journey, and making the best movie possible is the primary goal. Meanwhile, Adam Rogers at Wired has an in-depth look at the science behind the movie, including the (unsurprising, in retrospect) discovery that the super-accurate visualization software available to the Hollywood special-effects team enable the physicists to see things they hadn’t anticipated. Kip predicts that at least a couple of technical papers will come out of their work.

And that’s not all! Kip has a book coming out on the science behind the movie, which I’m sure will be fantastic. And there is also a documentary on “The Science of Interstellar” that will be shown on TV, in which I play a tiny part. Here is the broadcast schedule for that, as I understand it:

Wednesday, October 29, at 10pm PDT/9c

AHC (American Heroes Channel)
Sunday, November, 2 at 4pm PST/3c (with a repeat on Monday, November 3 at 4am PST/3c)

Thursday, November 6, at 11pm PST/10c

Of course, all the accurate science in the world doesn’t help if you’re not telling an interesting story. But with such talented people working together, I think some optimism is justified. Let’s show the world that science and cinema are partners, not antagonists.

by Sean Carroll at October 29, 2014 02:57 PM

Clifford V. Johnson - Asymptotia

Podcast on “Beyond the University”
That podcast from the Bedrosian Center series that I contributed to is now live! I mentioned it here, remember, and talked a bit about the book we discussed, "Beyond the University: Why Liberal Education Matters", by Michael S. Roth. In the discussion I chatted with the Center's Director Raphael Bostic, who also hosts, Deborah Natoli, and David Sloane, all in the USC Price School of Public Policy. It was a fun discussion, and I hope you find it interesting and useful. As I said in the previous post:
It is by no means a perfect book, as we discuss in the podcast, but it is in my view a book that is worth reading since it lays out rather nicely the history of the conversation that has been going on about this issue in America dating back to Jefferson and before. This is, to my mind, a conversation we will always need to have, an issue that is self-renewing and that has to be revisited, and we should all be part of it whether we are educators, parents, students, potential students, or employers. (Frankly, I think every new faculty member at a university that claims to be giving a liberal education should be given a copy of this book (or a book like it) in their arrival package. Existing faculty as well, if possible! Why? To get everyone involved in education thinking about the point of what it is they are doing, if nothing else.)
The podcast can be found on Soundcloud, iTunesU, and iTunes podcasts, but I'll point to it on the Center's page here since there are other [...] -cvj Click to continue reading this post

by Clifford at October 29, 2014 02:54 PM

arXiv blog

Google's Secretive DeepMind Startup Unveils a "Neural Turing Machine"

DeepMind has built a neural network that can access an external memory like a conventional Turing machine. The result is a computer that mimics the short-term memory of the human brain.


October 29, 2014 02:41 PM

arXiv blog

Google's Secretive DeepMind Start-up Unveils A "Neural Turing Machine"

DeepMind has built a neural network that can access an external memory like a conventional Turing machine. The result is a computer that mimics the short-term memory of the human brain.

One of the great challenges of neuroscience is to understand the short-term working memory in the human brain. At the same time, computer scientists would dearly love to reproduce the same kind of memory in silico.

October 29, 2014 02:41 PM

Christian P. Robert - xi'an's og

I am cold all over…

unusual snowfall on Bois de Boulogne, March 12, 2013An email from one of my Master students who sent his problem sheet (taken from Monte Carlo Statistical Methods) late:

Bonsoir Professeur
Je « suis » votre cours du mercredi dont le formalisme mathématique me fait froid partout
Avec beaucoup de difficulté je vous envoie mes exercices du premier chapitre de votre livre.

which translates as

Good evening Professor,
I “follow” your Wednesday class which mathematical formalism makes me cold all over. With much hardship, I send you the first batch of problems from your book.

I know that winter is coming, but, still, making students shudder from mathematical cold is not my primary goal when teaching Monte Carlo methods!

Filed under: Books, Kids, Statistics, University life Tagged: computational statistics, ENSAE, Master program, MCMC algorithms, Monte Carlo Statistical Methods, statistical computing, Université Paris Dauphine, Winter is coming

by xi'an at October 29, 2014 01:18 PM

astrobites - astro-ph reader's digest

Why Save the Sky?
Guest author Morgan Rehnberg, University of Colorado.

Guest author Morgan Rehnberg, University of Colorado.

This guest post is by Morgan Rehnberg, a graduate student in the Department of Astrophysical and Planetary Sciences at the University of Colorado – Boulder, where he studies the rings of Saturn. We met Morgan at the June 2014 ComSciCon national workshop, where this piece was first drafted. In his free time, he runs the popular-astronomy website Cosmic Chatter and advocates for increased public awareness of science. You can follow him on Twitter at @MorganRehnberg.

It’s been forty-five years since the world last gathered to look at the sky. That night, hundreds of millions of people would go outside to gaze at our moon and marvel at two men almost unfathomably far away. It was perhaps the final act in a romance which has spanned millennia: mankind and the night sky. Today, our knowledge of the cosmos dwarfs that of our ancestors, but our relationship with it is but a shadow of theirs. We’ve lost our connection to the Universe and, with it, a bit of ourselves. But, it’s not too late to reverse course. This breakup doesn’t have to be final, but we’ve got to act to save humanity’s first love.

Perhaps no event so highlights our disconnect from the sky as the 1994 Los Angeles earthquake. With power knocked out across the city, calls began flooding into emergency centers and television stations. Panicked citizens wanted to know if the earthquake had been caused by that “strange, giant, silvery cloud” hanging in the night sky. It wasn’t smoke or dust or clouds they were seeing. Thousands, perhaps millions, were seeing the Milky Way for the first time in their lives.

Before we enveloped ourselves in a cocoon of light, the stars used to dominate our nights. From a really dark place, you can see perhaps five thousand individual stars. Billions more lend their glow to the Milky Way. Today, there are but a few places in the developed world where one can take in such a view. Prior to the invention of electric lighting, there was practically nowhere you couldn’t.

Our ancestors depended upon the night sky. They navigated by the stars to circle the globe and to escape from slavery. They planted their crops in accordance with the appearance of certain objects and marked time by observing the solstices. The stars weren’t just pretty; they were invaluable tools.

To think of the sky as merely the first calendar, however, is to vastly underestimate its psychological impact. Virtually every culture on Earth told stories set against the backdrop of the cosmos and formed the bright stars into their characters. The tales are as varied as the civilizations which envisioned them. Our modern constellations may be derived from those of Greece and Rome, but every empire had different ideas. Perhaps most remarkable are the constellations of some cultures in the Southern Hemisphere. There the sky was so dark and the Milky Way so brilliant that peoples like the Inca formed pictures not by connecting stars, but by looking for the gaps. Mach’acuay, Lord of All Serpents, for example, occupied one of the great dust lanes of our galaxy. This dust extinguishes the light of stars and leaves behind a striking expanse of darkness. It was this void that formed the great snake.

The canvas could not endure forever. In 1831, Michael Faraday became the first person to generate electricity. His discovery would change the world, but it would start an inexorable path towards our isolation from the rest of nature. It’d be foolish, though, to criticize our forefathers for choosing the path they did; electricity enabled an unprecedented leap in the powers of mankind. With it we learned to communicate across vast distances, extend our homes towards the heavens, and fight the diseases that haunted us. Electricity changed us for the better.

But, as town after town began to light its streets, electricity also allowed us to beat back the night sky. Light became a badge of progress. From New York to London to Tokyo, night turned into a second day. The intricate tapestry of the Milky Way faded into a pale gray blur. Stars gave way to skyscrapers. As the march of progress hastened, so did the fading of our cultural memories. The night sky became like the American Dream: an idea we all acknowledged was out there, lurking just beyond our grasp.

You may ask, does all of this really matter? With billions living in poverty and our climate irrevocably changing, should we bother fretting about a few lost stars? The answer must be yes. Today, our priorities have never been more provincial. We’re more concerned about the next financial quarter or upcoming film release than the long-term health of our people, societies, and planet. Maybe it shouldn’t surprise us that we increasingly focus on ourselves when we’ve blocked out the rest of the Universe. Would people still refuse to vaccinate their children or fund science or feed the needy if they could see the Milky Way? Of course. But would a nightly, visceral reminder of our place in the cosmos force some to think more broadly? I think so. The Earth is a tiny vessel adrift in an enormous sea; perhaps the universal experience of a brilliant sky would remind us that we have more to bind us together than to cleave us apart.

Distressingly, the effects of losing our night sky may be even more insidious than simply changing minds. A brighter night may be changing our bodies, as well, and not for the better. Evidence is beginning to mount that exposure to artificial light at night changes the very chemistry of our organs. Cancer, chronic fatigue, and more, have been attributed to these subtle shifts. As people around the world flock to cities, this hidden epidemic will grow. Thankfully, the cure may be within our reach: restoring a dark night so that our bodies may rest as they have for millions of years.

It’s not too late for us to take action. It’s easier than you might think. Start small: if you use outdoor lighting for your home, consider removing it. If you deem such fixtures a necessity, ensure that they all direct light downwards. Not only will this prevent stray light from escaping to the sky, it will also better illuminate your property. Contact your local government and encourage them to do the same. Downwards-directed lighting can substantially reduce the electric bill for any institution. Need some inspiration? Visit Flagstaff, AZ, the world’s first International Dark Sky Community. I’ve been there and, let me tell you, walking out of a restaurant to the sight of the Milky Way is an astonishing experience.

What about safety? Don’t we need those lights to protect our families, friends, and neighbors? It’s true, we do – to an extent. Beyond a base level, however, more light simply does not correlate with less reported crime. Bright light cuts two ways, helping the criminal just as much as the homeowner. Modern communities need light at night, but only the right light. Any more than that and we’re just shoveling money down the drain.

We can do this. One at a time, street by street, town by town, we can save the sky. For thousands of years, our view of the Universe inspired us to do the magnificent. Let’s take that back for future generations, because, as our grasp widens, so must our appreciation for the cosmos into which we reach.

by Guest at October 29, 2014 12:04 PM

Peter Coles - In the Dark

Three Minutes of Cosmology

Not much time today to do anything except help one of my former PhD students become a Youtube sensation by sharing this video of Ian Harrison. Ian did his doctorate with me in Cardiff but now works in the Midlands, at the University of Manchester. Here he is talking about part of his PhD work for just three minutes without repetition, hesitation, deviation, or repetition:



by telescoper at October 29, 2014 11:01 AM

Lubos Motl - string vacua and pheno

Maldacena's fairy-tale on exchange rates, gauge theories, and the Higgs
Juan Maldacena wrote a popular essay on gauge symmetries and their breaking:
The symmetry and simplicity of the laws of physics and the Higgs boson (PDF).
It's sometimes being said that Einstein's discovery of the special theory of relativity was helped by Einstein's work with the patents dealing with the train synchronization. I think that it's not really true (Einstein was stunned by the apparent contradiction between Maxwell's field theory and Newtonian mechanics long before he worked with the patents: sorry, Peter Galison) but it's good enough as a fairy-tale.

The next Maldacena may arrive from Zimbabwe.

Analogously, we learn that Maldacena's work with gauge theories was helped by the chronic inflation in his homeland, Argentina. The persistent monetary inflation and currency reforms – something that many of us consider to be "once in a century" painful event – became as mundane in Argentina as a gauge transformation. In fact, as Maldacena shows (and he is not the first one, I guess), it is not just an analogy. The switching to another unit of wealth is a special case of a gauge transformation.

With this experience, a European or North American gauge theorist facing Maldacena must feel just like a European soccer player facing Argentina, if we recall another observation by Juan at Strings 2014.

The paper is full of images from Wikipedia. The beginning is all about the Beauty and the Beast and the concept of symmetry. You are also reminded about the electricity and magnetism.

But the financial representation of the gauge field \(A_\mu\) and the gauge symmetry is the most interesting thing, of course. The financial gauge group is isomorphic to \(\RR\) but otherwise it works well. Maldacena offers you a financial interpretation of the field strength \(F_{\mu\nu}\) as well, of course.

If you think about a lattice version of the gauge theory, the links correspond to a "conversion of one currency from another". If you go around a loop constructed from these links, the exchange rates may be mismatched and you may earn (or lose) money. Your original assets ("before" you make the round trip) get pretty much multiplied by a factor\[

{\rm Assets}_{\rm after} = {\rm Assets}_{\rm before} \cdot \exp(\text{Magnetic flux})

\] if I add a formula to Maldacena's presentation. The exponential is the monodromy, the Wilson loop without any trace. You shouldn't forget that the trading gauge group is noncompact.

In the real world, we have to consider all these objects in the quantum mechanical language, as Maldacena discusses in another section, and the too obviously wrong consequences of the most naive realization of the symmetry must be avoided by the spontaneous symmetry breaking, the Higgs mechanism.

If someone likes semi-technical popular texts on physics, I recommend it.

by Luboš Motl ( at October 29, 2014 06:58 AM

October 28, 2014

The Great Beyond - Nature blog

Private rocket explodes on launch to space station
Flames engulfed the rocket seconds after lift-off.

Flames engulfed the rocket moments after lift-off.


An Orbital Sciences Antares rocket exploded seconds after its 6:22 p.m. lift-off from Wallops Island, Virginia, Tuesday on a mission to resupply the International Space Station. No one was hurt, but the rocket was apparently destroyed and there was “significant property damage”, according to mission control commentators on NASA television.

“We have lost the vehicle,” said controllers from the Johnson Space Center in Houston. “The [space station] crew has been informed of the accident.”

Orbital moved almost immediately into contingency mode, asking its engineers to retain all notes and photographs related to the launch. Fires could be seen burning across the launch pad. “Obviously we will need to instigate an accident investigation team,” the launch director said.

Orbital, of Dulles, Virginia, is one of two private companies flying cargo to the space station for NASA. Its competitor, SpaceX of Hawthorne, California, is aiming to eventually carry astronauts as well.

It was the third of eight scheduled missions for Orbital. Among the 2,300 kilograms of cargo on board were a spectrometer to measure meteors entering the atmosphere and a neck collar for astronauts to measure blood flow from the brain. The payload also included test hardware for a future private asteroid prospecting mission, as well as unspecified classified cryptography equipment.

Launch of a Russian Progress vehicle, scheduled for the morning of 29 October with more crew supplies, was not expected to be affected.


by Alexandra Witze at October 28, 2014 11:29 PM

Christian P. Robert - xi'an's og

reliable ABC model choice via random forests

human_ldaAfter a somewhat prolonged labour (!), we have at last completed our paper on ABC model choice with random forests and submitted it to PNAS for possible publication. While the paper is entirely methodological, the primary domain of application of ABC model choice methods remains population genetics and the diffusion of this new methodology to the users is thus more likely via a media like PNAS than via a machine learning or statistics journal.

When compared with our recent update of the arXived paper, there is not much different in contents, as it is mostly an issue of fitting the PNAS publication canons. (Which makes the paper less readable in the posted version [in my opinion!] as it needs to fit the main document within the compulsory six pages, relegated part of the experiments and of the explanations to the Supplementary Information section.)

Filed under: pictures, R, Statistics, University life Tagged: 1000 Genomes Project, ABC, ABC model choice, machine learning, model posterior probabilities, posterior predictive, random forests, summary statistics

by xi'an at October 28, 2014 11:14 PM

Emily Lakdawalla - The Planetary Society Blog

[Update 2] Antares Rocket Explodes Seconds after Liftoff
An Antares rocket fell back to the launch pad shortly after liftoff, exploding in a fireball that destroyed the vehicle.

October 28, 2014 11:13 PM

Clifford V. Johnson - Asymptotia

Interstellar Discoveries
I'm a fan of Chris Nolan's work so I've been looking forward to Interstellar. I've also been fascinated by the McConaussance - the transformation of Matthew McConaughey into an actor of considerable stature in a series of excellent films (Mud, Dallas Buyers Club, etc...), so I've been doubly interested in seeing how he works in a film under Nolan's direction. Same for the always amazing Casey Affleck. All quite exciting to see. But then to my surprise it turns out there's another reason to be interested. Kip Thorne. Some years ago, at a party when I last saw him, Kip told me that he had been working on some film or other with a major studio, but I did not know of the details. Then I ran into a mutual friend a couple of months ago who said something a long the lines of "Kip's movie is coming out soon...", and I learned that it was something to do with Interstellar! But I did not know any details. Then I got sent* this Wired story, and then** this story, and I finally got around to looking. The Wired story has a lot of interesting detail, including a special film (that I ought to look at at) with interviews and behind the scenes material (the still to the right is a screen shot from it). still_from_interstellar_wiredThe film will apparently feature a black hole and a wormhole in some way (I don't want to know more - I like films to unfold in front of me in the theatre). Kip has been working with the visual effects people to get right exactly how such objects really look, an issue that has not really been fully addressed, it seems. He, like a number of us interested in science and film, is keen to help filmmakers really do a good job of representing some of these fascinating objects as accurately as possible. (Not, in my view, in order to stifle filmmakers' imagination, as it so often seems when you hear scientists out there pontificating about what's wrong in one film or another, but because the actual science is so very often far more interesting and full of delights and possibility than a visual effects kluge can be...) So apparently he wrote down [...] Click to continue reading this post

by Clifford at October 28, 2014 04:30 PM

Symmetrybreaking - Fermilab/SLAC

Scientists mull potential gamma-ray study sites

An international panel is working to determine the two locations from which the Cherenkov Telescope Array will observe the gamma-ray sky.

Somewhere in the Southern Hemisphere, about 100 state-of-the-art telescopes will dot the otherwise empty landscape for half a kilometer in every direction. Meanwhile, in the Northern Hemisphere, a swath of land a little over a third the size will house about 20 additional telescopes, every one of them pointing toward the heavens each night for a full-sky view of the most energetic—and enigmatic—processes in the universe. 

This is the plan for the Cherenkov Telescope Array Observatory, the world’s largest and most sensitive gamma-ray detector. The construction of the first of the two arrays is scheduled to begin in 2016, with the observatory becoming fully operational by 2020. At that point, CTA’s telescopes will observe gamma rays produced in some of the universe’s most violent events—everything from supernovas to supermassive black holes. 

Yet where exactly the telescopes will be built remains to be seen.

Scientists representing the 29-country CTA consortium met last week to discuss the next steps toward narrowing down potential sites in the Northern Hemisphere: two in the United States (both in Arizona) and two others in Mexico and the Canary Islands. Although details from that meeting remain confidential, the CTA resource board is expected to begin negotiations with the potential host countries within the next few months. That will be the final step before the board makes its decision, says Rene Ong, co-spokesperson of CTA and a professor of physics and astronomy at UCLA.

“Whichever site it goes to, it will be very important in that country,” Ong says. “It’s a major facility, and it will bring with it a huge amount of intellectual capital.”

Site selection for the Southern Hemisphere is a bit further along. Last April, the CTA resource board narrowed down that list to two potential sites: one in Southern Namibia and one in Northern Chile. The board is now in the process of choosing between the sites based on factors including weather, operating costs, existing infrastructure like roads and utilities, and host country contributions. A final decision is expected soon.

Artwork by: Sandbox Studio, Chicago

“The consortium went through an exhaustive 3-year process of examining the potential sites, and all of the sites now being considered will deliver on the science,” says CTA Project Scientist Jim Hinton, a professor of physics and astronomy at the University of Leicester. “We’re happy that we have so many really good potential sites. If we reach an impasse with one, we can still keep moving forward with the others.”

Scientists do not completely understand how high-energy gamma rays are created. Previous studies suggest that they stream from jets of plasma pouring out of enormous black holes, supernovae and other extreme environments, but the processes that create the rays—as well as the harsh environments where they are produced—remain mysterious.

To reach its goal of better understanding high-energy gamma rays, CTA needs to select two sites—one in the Northern Hemisphere and one in the Southern Hemisphere—to see the widest possible swath of sky. In addition, the view from the two sites will overlap just enough to allow experimenters to better calibrate their instruments, reducing error and ensuring accurate measurements.

With 10 times the sensitivity of previous experiments, CTA will fill in the many blank regions in our gamma-ray map of the universe. Gamma-rays with energies up to 100 gigaelectronvolts have already been mapped by the Fermi Gamma-ray Space Telescope and others; CTA will cover energies up to 100,000 gigaelectronvolts. It will survey more of the sky than any previous such experiment and be significantly better at determining the origin of each gamma ray, allowing researchers to finally understand the astrophysical processes that produce these energetic rays.

CTA may also offer insight into dark matter. If a dark matter particle were to naturally decay or interact with its antimatter partner to release a flash of energy, the telescope array could theoretically detect that flash. In fact, CTA is one of very few instruments that could see such flashes with energies above 100 gigaelectronvolts.  

“I’m optimistic that we’ll see something totally new and unexpected,” Ong says. “Obviously I can’t tell you what it will be—otherwise it wouldn’t be unexpected—but history tells us that when you make a big step forward in capability, you tend to see something totally new. And that’s just what we’re doing here.”


Like what you see? Sign up for a free subscription to symmetry!

by Kelen Tuttle at October 28, 2014 03:42 PM

Tommaso Dorigo - Scientificblogging

Another Collider Physics Source
Just a short entry to mention that the blog of my colleague Michael Schmitt, a professor at Northwestern University and a member of the CMS collaboration, is as active as ever, with several very interesting and well-written pieces recently published:

read more

by Tommaso Dorigo at October 28, 2014 03:00 PM

astrobites - astro-ph reader's digest

Pair-Instability Supernovae: What might they look like?

What happens to the most massive stars?

Really large stars eventually collapse because there is not enough radiation pressure to prevent the outer layers from falling in gravitationally. These stars then explode as Type II supernovae. Even more massive stars, between about 130 and 250 solar masses, are thought to lead to a pair-instability supernovae (PISN). In these stars, electron-positron pairs are created in the core. This leads the star to become dynamically unstable and leads to the collapse then explosion of these stars. This recent Astrobite has a more thorough explanation of the physics of PISN. These types of supernovae should be extremely luminous, much more than other Type II supernovae. At least, this is theoretical prediction: a pair-instability supernova has never been observed, though there have been a few candidates.

There have been previous models of PISN that show these stars evolve onto the Red-SuperGiant Branch before collapsing. However, these models have left out an important aspect of Red SuperGiants. Often, the hydrogen-rich outer envelope of the Red SuperGiant will begin pulsating. If the ratio of the luminosity of the star to the stellar mass is large, these pulsations can become unstable and lead to significant mass loss. The authors investigate how these pulsations evolve over time. More specifically, they want to know if these pulsations can lead to the star losing mass and if so, how this would impact observations of these type of supernovae.

What does these new models tell us?

Figure 1: The radius and mass-loss rate of multiple stellar models over 200 years. Epsilon is the fraction of the kinetic energy gained from the pulsations that contributes to material from the star being expelled from the surface. The top plot shows the radius as a function of time, but can also be thought of as showing how the pulsations evolve over time. The y-axis on the bottom plot is the log of the amount of mass lost per year, measured in solar masses. Manner A (Black) shows an exponential increase in the amplitude, then constant. Manner B (Brown and Cyan) show a quick increase in the pulsations followed by damping. Manner C (Red and Navy Blue) show continued exponential growth of the pulsations.

The authors used a code called MESAstar to model the evolution of three stars of 150, 200, and 250 solar masses. Previous studies of pulsations in Red SuperGiant stars have found pulsational periods around 1000 days. Therefore, the authors chose timesteps for the models of under a day to understand how these pulsations would evolve. They find the pulsations can evolve in three different manners. In manner A, the amplitude of the pulsations grow exponentially, but then saturate and continue at a constant amplitude. In manner B, the amplitude grows exponentially before suddenly being damped out. In manner C, the amplitudes grow exponentially until material on the stellar surface begins to move at the escape velocity.

Now that the authors know how these pulsations evolve in Red SuperGiants, they seek to understand how this can lead to mass loss from the star. When material on the stellar surface gains enough kinetic energy from the pulsations, it can escape the star’s gravity and be expelled into the surrounding space. Figure 1 shows how the pulsations evolve over time (top) and how the rate of mass loss changes with time (bottom). The different curves correspond to different fractions of the total kinetic energy from the pulsations that are used to enhance the mass loss.

The evolution of the pulsations affects the mass loss rate. The black curve shows what happens when the pulsations increase then become constant (A). Brown and Cyan match repeated growth and damping of the amplitudes (B). Red and Navy Blue correspond to when the amplitude continues to grow exponentially (C). All of these models show the pulsations can lead to a mass-loss rate of around 0.01 solar masses per year. This is significant compared to previous results, which assumed zero mass lass, and changes what astronomers expect to see observationally.

How can astronomers test these results?

If PISN are losing significantly more mass than previously thought, the environment around the star will be very dense. The presence of this material will change what astronomers observe if they ever see a PISN. This environment will cause these supernovae to be bluer and brighter than previously thought. As the energetic ejected material from the supernova hits the surrounding material, this material is heated and photoionized, which causes more brightening. It has long been thought that the earliest supernovae in the universe did not suffer much mass loss and exploded in regions without much circumstellar material. This result means it should be possible to observe these supernovae at higher redshifts than astronomers thought, which has implications for determining distances in the universe.

by Josh Fuchs at October 28, 2014 02:19 PM

ATLAS Experiment

Defending Your Life (Part 3)

This is the last part of my attempt to explain our simulation software. You can read Part 1, about event generators, and Part 2, about detector simulation, if you want to catch up. Just as a reminder, we’re trying to help our theorist friend by searching for his proposed “meons” in our data. The detector simulation gives us a long list of energy deposits, times, and locations in our detector. The job isn’t done though. Now we have to take those energy deposits and turn them into something that looks like our data – which is pretty tricky! The code that does that is called “digitization”, and it has to be written specially for our detector (CMS has their own too).

The simple idea is to change the energies into whatever it is that the detector reads out – usually times, voltages, and currents, for example, but it can be different for each type of detector. We have to build in all the detector effects that we care about. Some are well known, but not well understood (Birk’s law, for example). Some are a little complicated, like the change in light collected from a scintillator tile in the calorimeter depending on whether the energy is deposited right in the middle or on the edge. We can use the digitization to model some of the very low-energy physics that we don’t want to have to simulate in detail with Geant4 but want to get right on average. Those are effects like the spread and collection of charge in a silicon module or the drift of ionized gas towards a wire at low voltage.

Z to mu mu at high pileup

One of our events with lots of “pile-up” – many standard proton-proton collisions, one dot for each, on top of one that we’re interested in (the one with the yellow tracks)

Digitization is where some other effects are put in, like “pile-up“, which is what we call the extra proton-proton collisions in a single bunch crossing. Those we usually pre-simulate and add on top of our important signal (meon) events, like using a big library. We can add other background effects if we want to, like cosmic rays crossing the detector, or proton collisions with remnant gas particles floating around in the beampipe, or muons running down parallel to the beamline from protons that hit collimators upstream. Those sorts of things don’t happen every time protons collide, but we sometimes want to study how they look in the detector too.

Now we should have something that looks a lot like our data – except we know exactly what it is, without any ambiguity! With that, we can try to figure out if our friend’s meons are a part of nature. We can build up some simulated data that includes all the different processes that we already know exist in nature, like the production of top quarks, W bosons, Z bosons, and our new Higgs bosons. And we can build another set that has all of those things, but also includes our friend’s meons. The last part, which is really what our data analysis is all about, is trying to figure out what makes events with meons special – different from the other ones we expect to see – and trying to isolate them from the others. We can look at the reconstructed energy in the event, the number of particles we find, any oddities like heavy particles decaying away from the collision point – anything that helps. And we have to know a little bit about the simulation, so that we don’t end up using properties of the events that are very hard to get right in the simulation to separate meons from other particles. That really is the first part of almost all our data analyses. And the last part of most of our analyses (we hope), is “unblinding”, where we finally check the data that has all the features we want – passes all our requirements – and see whether it looks more like nature with or without meons. Sometimes we try to use “data-driven methods” to estimate the backgrounds (or tweak the estimates from our simulation), but almost every time we start from the simulation itself.

Some of our data with a few different guesses as to what new physics might look like (here different dark matter models). The data look more like our expectation without them, though – so no new physics today!

The usual thing that we find is that our friend told us about his theory, and we looked for it and didn’t find anything exciting. But by the time we get back, our theorist friends often say “well, I’ve been thinking more, and actually there is this thing that we could change in our model.” So they give you a new version of the meon theory, but this time instead of being just one model, it’s a whole set of models that could exist in nature, and you have to figure out whether any of them are right. We’re just going through this process for Supersymmetry, trying to think of thousands of different versions of Supersymmetry that we could look for and either find or exclude. Often, for that, you want something called a “fast simulation.”

To make a fast simulation, we either go top-down or bottom-up. The top-down approach means that we look at what the slowest part of our simulation is (always the calorimeters) and find ways to make it much, much faster, usually by parameterizing the response instead of using Geant4. The bottom-up approach means that we try to skip detector simulation and digitization all together and go straight to the final things that we would have reconstructed (electrons, muons, jets, missing transverse momentum). There are even public fast simulations like DELPHES and the Pretty Good Simulation that theorists often use to try to find out what we’ll see when we simulate their models. Of course, the faster the simulation, normally, the fewer details and oddities can be included, and so the less well it models our data (“less well” doesn’t have to be “not good enough”, though!). We have a whole bunch of simulation types that we try to use for different purposes. The really fast simulations are great for quickly checking out how analyses might work, or for checking out what they might look like in future versions of our detector in five or ten years.

So that’s just about it – why we really, really badly need the simulation, and how each part of it works. I hope you found it a helpful and interesting read! Or at least, I hope you’re convinced that the simulation is important to us here at the LHC.

ZachMarshall Zach Marshall is a Divisional Fellow at the Lawrence Berkeley National Laboratory in California. His research is focused on searches for supersymmetry and jet physics, with a significant amount of time spent working on software and trying to help students with physics and life in ATLAS.

by Zachary Marshall at October 28, 2014 12:30 PM

Peter Coles - In the Dark

STFC Consolidated Grants Review

It’s been quite a while since I last put my community service hat on while writing a blog post, but here’s an opportunity. Last week the Science and Technology Facilities Council (STFC) published a Review of the Implementation of Consolidated Grants, which can be found in its entirety here (PDF). I encourage all concerned to read it.

Once upon a time I served on the Astronomy Grants Panel whose job it was to make recommendations on funding for Astronomy through the Consolidated Grant Scheme, though this review covers the implementation across the entire STFC remit, including Nuclear Physics, Particle Physics (Theory), Particle Physics (Experiment) and Astronomy (which includes solar-terrestrial physics and space science). It’s quite interesting to see differences in how the scheme has been implemented across these various disciplines, but I’ll just include here a couple of comments on the Astronomy side of things.

First, here is a table showing the number of academic staff for whom support was requested over the three years for which the consolidated grant system has been in existence (2011, 2012 and 2013 for rounds 1, 2 and 3 respectively).  You can see that the overall success rate was slightly better in round 3, possibly due to applicants learning more about the process over the cycle, but otherwise the outcomes seem reasonably consistent:


The last three rows of this table  on the other hand show quite clearly the impact of the “flat cash” settlement for STFC science funding on Postdoctoral Research Assistant (PDRA) support:

Constant cash means ongoing cuts in real terms; there were 11.6% fewer Astronomy PDRAs supported in 2013 than in 2011. Job prospects for the next generation of astronomers continue to dwindle…

Any other comments, either on these tables or on the report as a whole, are welcome through the comments box.


by telescoper at October 28, 2014 12:05 PM

Lubos Motl - string vacua and pheno

Distance between quantum field theories
The first and most interesting hep-th paper today is
Relative Entropy and Proximity of Quantum Field Theories
by Vijay Balasubramanian, Jonathan J. Heckman, and Alexander Maloney. They use the notion of relative entropy\[

D_{KL} (p||q) = \int \dd\mu (z)\, p(z)\log \frac{p(z)}{q(z)}

\] that is well-known to folks in machine learning (Kaggle anyone?) and related fields to quantify how far two quantum field theories are, how much information you lose when you run from the ultraviolet to the infrared.

Their formalism generalizes the Zamolodčikov metric on the space of couplings – that special case is only applicable if the field theories are conformal. (It's funny to write about these matters because I know Vijay, Jonathan, Alex, as well as Saša Zamoldčikov very well – and yes, Saša liked the Czech spelling of his name when I suggested it to him LOL.)

These formulae may be used to quantify the amount of fine-tuning in field theories in a new way.

But the main reason why I think that this kind of research is important, and I am doing something similar as well, is that similar "measures" may be applied to the string landscape and may be used to define rules of vacuum selection that Nature may have used Herself.

In particular, for many years, I have believed in something that I sometimes called the "misanthropic principle". The vacua in the "regions" where vacua are dense – where the average distance to tne nearest neighbor is tiny – must be disfavored, along with all the "highly populous classes" of the vacua that the anthropic believers tend to favor. In other words, we should live in some of the vacua that are "most special" and "most separated" from all the neighbors, according to the appropriate measure.

It's been my belief that such a principle or "bias" is needed for the probability distribution (saying which vacua are likely as initial conditions for the evolution of the Universe) to be normalizable.

This principle, if true, could perhaps be derived from some rather natural mathematics, some Hartle-Hawking-like wave function on the stringy configuration space or something like that. The wave function would drop as you would move towards the "dense, highly populated regions" of the landscape – for similar reasons as the reasons why the harmonic oscillator ground state wave function decreases for large values of \(x\) or why the Boltzmann distribution decreases if the energy \(E\) increases.

by Luboš Motl ( at October 28, 2014 06:42 AM

October 27, 2014

Christian P. Robert - xi'an's og

projective covariate selection

While I was in Warwick, Dan Simpson [newly arrived from Norway on a postdoc position] mentioned to me he had attended a talk by Aki Vehtari in Norway where my early work with Jérôme Dupuis on projective priors was used. He gave me the link to this paper by Peltola, Havulinna, Salomaa and Vehtari that indeed refers to the idea that a prior on a given Euclidean space defines priors by projections on all subspaces, despite the zero measure of all those subspaces. (This notion first appeared in a joint paper with my friend Costas Goutis, who alas died in a diving accident a few months later.) The projection further allowed for a simple expression of the Kullback-Leibler deviance between the corresponding models and for a Pythagorean theorem on the additivity of the deviances between embedded models. The weakest spot of this approach of ours was, in my opinion and unsurprisingly, about deciding when a submodel was too far from the full model. The lack of explanatory power introduced therein had no absolute scale and later discussions led me to think that the bound should depend on the sample size to ensure consistency. (The recent paper by Nott and Leng that was expanding on this projection has now appeared in CSDA.)

“Specifically, the models with subsets of covariates are found by maximizing the similarity of their predictions to this reference as proposed by Dupuis and Robert [12]. Notably, this approach does not require specifying priors for the submodels and one can instead focus on building a good reference model. Dupuis and Robert (2003) suggest choosing the size of the covariate subset based on an acceptable loss of explanatory power compared to the reference model. We examine using cross-validation based estimates of predictive performance as an alternative.” T. Peltola et al.

The paper also connects with the Bayesian Lasso literature, concluding on the horseshoe prior being more informative than the Laplace prior. It applies the selection approach to identify biomarkers with predictive performances in a study of diabetic patients. The authors rank model according to their (log) predictive density at the observed data, using cross-validation to avoid exploiting the data twice. On the MCMC front, the paper implements the NUTS version of HMC with STAN.

Filed under: Mountains, pictures, Statistics, Travel, University life Tagged: Aki Vehtari, Bayesian lasso, Dan Simpson, embedded models, Hamiltonian Monte Carlo, horseshoe prior, Kullback-Leibler divergence, MCMC, Norway, NUTS, predictive power, prior projection, STAN, variable selection, zero measure set

by xi'an at October 27, 2014 11:14 PM

John Baez - Azimuth

Biodiversity, Entropy and Thermodynamics


I’m giving a short 30-minute talk at a workshop on Biological and Bio-Inspired Information Theory at the Banff International Research Institute.

I’ll say more about the workshop later, but here’s my talk, in PDF and video form:

Biodiversity, entropy and thermodynamics.

Most of the people at this workshop study neurobiology and cell signalling, not evolutionary game theory or biodiversity. So, the talk is just a quick intro to some things we’ve seen before here. Starting from scratch, I derive the Lotka–Volterra equation describing how the distribution of organisms of different species changes with time. Then I use it to prove a version of the Second Law of Thermodynamics.

This law says that if there is a ‘dominant distribution’—a distribution of species whose mean fitness is at least as great as that of any population it finds itself amidst—then as time passes, the information any population has ‘left to learn’ always decreases!

Of course reality is more complicated, but this result is a good start.

This was proved by Siavash Shahshahani in 1979. For more, see:

• Lou Jost, Entropy and diversity.

• Marc Harper, The replicator equation as an inference dynamic.

• Marc Harper, Information geometry and evolutionary game theory.

and more recent papers by Harper.

by John Baez at October 27, 2014 09:56 PM

Peter Coles - In the Dark

Poets in October

I’m sure few readers of this blog can have failed to notice that today is the 100th anniversary of the birth of Welsh poet Dylan Thomas. I’ve posted quite a few of his poems over the years so it seems fitting to post this, Poem in October, as a birthday tribute.

It was my thirtieth year to heaven
Woke to my hearing from harbour and neighbour wood
And the mussel pooled and the heron
Priested shore
The morning beckon
With water praying and call of seagull and rook
And the knock of sailing boats on the webbed wall
Myself to set foot
That second
In the still sleeping town and set forth.

My birthday began with the water-
Birds and the birds of the winged trees flying my name
Above the farms and the white horses
And I rose
In a rainy autumn
And walked abroad in shower of all my days
High tide and the heron dived when I took the road
Over the border
And the gates
Of the town closed as the town awoke.

A springful of larks in a rolling
Cloud and the roadside bushes brimming with whistling
Blackbirds and the sun of October
On the hill’s shoulder,
Here were fond climates and sweet singers suddenly
Come in the morning where I wandered and listened
To the rain wringing
Wind blow cold
In the wood faraway under me.

Pale rain over the dwindling harbour
And over the sea wet church the size of a snail
With its horns through mist and the castle
Brown as owls
But all the gardens
Of spring and summer were blooming in the tall tales
Beyond the border and under the lark full cloud.
There could I marvel
My birthday
Away but the weather turned around.

It turned away from the blithe country
And down the other air and the blue altered sky
Streamed again a wonder of summer
With apples
Pears and red currants
And I saw in the turning so clearly a child’s
Forgotten mornings when he walked with his mother
Through the parables
Of sunlight
And the legends of the green chapels

And the twice told fields of infancy
That his tears burned my cheeks and his heart moved in mine.
These were the woods the river and the sea
Where a boy
In the listening
Summertime of the dead whispered the truth of his joy
To the trees and the stones and the fish in the tide.
And the mystery
Sang alive
Still in the water and singing birds.

And there could I marvel my birthday
Away but the weather turned around. And the true
Joy of the long dead child sang burning
In the sun.
It was my thirtieth
Year to heaven stood there then in the summer noon
Though the town below lay leaved with October blood.
O may my heart’s truth
Still be sung
On this high hill in a year’s turning.


I admire greatly the poems of Dylan Thomas for their energy and colour and the truly original way he uses words. Nevertheless I do agree with a friend of mine who said earlier today that Dylan Thomas is the second-greatest Welsh poet of the twentieth century who wrote in English and had the surname “Thomas”. I mean no disrespect at all to DM but, although the two are very different, RS has to be the greater of the two Thomases.

This poem by R.S. Thomas is called Song at a Year’s Turning; the echo of the final phrase of Poem in October and the fact that it was published in 1955 make it very clear that it was written as a kind of elegy from one Thomas to the other:

Shelley dreamed it. Now the dream decays.
The props crumble; the familiar ways
Are stale with tears trodden underfoot.
The heart’s flower withers at the root.
Bury it then, in history’s sterile dust.
The slow years shall tame your tawny lust.

Love deceived him; what is there to say
The mind brought you by a better way
To this despair? Lost in the world’s wood
You cannot stanch the bright menstrual blood.
The earth sickens; under naked boughs
The frost comes to barb your broken vows.

Is there blessing? Light’s peculiar grace
In cold splendour robes this tortured place
For strange marriage. Voices in the wind
Weave a garland where a mortal sinned.
Winter rots you; who is there to blame?
The new grass shall purge you in its flame.

by telescoper at October 27, 2014 05:29 PM

Lubos Motl - string vacua and pheno

Tiny dark energy from co-existence of phases
There are some interesting hep-ph articles on the arXiv today. I will mention two of them – both papers are available on owned by Elsevier (so their title pages are almost identical). First, there is the paper
SUSY fits with full LHC Run I data
by the MasterCode Collaboration (represented by Kees Jan de Vries). They present the state-of-the-art fits on simple supersymmetric models. Unless one relies on a significant amount of good luck, the anomalous value of the muon magnetic moment can only be satisfactorily explained by the superpartners – without contradicting the current bounds imposed by the LHC – if one sacrifices the usual grand-unification-inspired relationship between the MSSM couplings.

But I want to spend more time with
On the smallness of the cosmological constant
by Froggatt, Nevzorov, Nielsen, and Thomas (they have written similar papers in the past). They want to explain the small value of the cosmological constant – and perhaps also of the Higgs mass – in the Planck units using their "Multiple Point Principle" (MPP).

The principle first postulated in 1993 or 1995 says that Nature wants to choose the values of some parameters in the theory to allow the co-existence of "phases" with the same value of the energy density. In this particular paper, they study the principle in the context of supergravity models that allow the co-existence of supersymmetry-preserving and supersymmetry-breaking Minkowski (zero cosmological constant) vacua.

The supersymmetry-preserving vacua may have demonstrably vanishing (or tiny?) values of the cosmological constant (or the Higgs mass), and the MPP then implies the same virtues for the supersymmetry-breaking vacua (for which the vanishing or smallness otherwise looks like fine-tuning).

It seems to me that they only use the other vacua to explain the properties of our vacuum – while the other degenerate vacua remain "unused in physics". In this sense, if I understood the paper well, they are not proposing a version of my favorite paradigm of the cosmological constant seesaw mechanism. I am still in love with the latter idea, spending some research time with it. Sometime in the future, I may write some update about it – how it effectively deals with the existence of fields that take values in a discrete set and how calculations may be done with such variables.

These authors are linking the small magnitude of the dark energy and the Higgs boson's lightness to the vanishing of some couplings near the Planck scale which also seems to be equivalent to the bounds on the Higgs stability and to the degeneracy of the vacua, and so on. There are some specific computations concerning the gaugino condensates etc.

The tip of an iceberg swimming in the ocean is likely to be close to 0 °C.

Unfortunately, the principles and logic of the new way of looking at these things aren't spelled out too clearly. In particular, I don't understand whether MPP could be compatible with string theory. String theory dictates the value of couplings in all of its effective field theories. But could the stringy ways to calculate them be (approximately or exactly) equivalent to the MPP, the requirement that phases may co-exist? Why would it work in this way? And if this doesn't work exactly in this way, could there be some room for such a new principle adjusting the effective couplings?

Why should the MPP be true in the first place? In the usual reasoning about naturalness, we allow some otherwise unnatural values of couplings if they are able to increase the amount of symmetry of the physical theory. A theory with a larger symmetry is "qualitatively different", so the probability distribution may have a delta-function-like peak at the loci of the parameter space with enhanced symmetry.

But maybe points of the parameter space that allow the "co-existence" (can they really co-exist!?) of two ore several phases are also "qualitatively different", even if there is no obvious enhanced symmetry, so they may be more likely. Is that what we should believe? And if it is, is there a clearer justification for this belief?

The analogy with the phase transitions of solids, liquids, and gases may be highly attractive. Think about water. The temperature \(T\) is a continuous parameter and all values are equally likely. In particular, you might predict that the percentage of water whose temperature is 0 °C (plus minus 0.01 °C) will be tiny.

Except that when ice is melting or water is freezing, it spends some time exactly at 0 °C (or whatever is the relevant melting point) because heat isn't being spent to change \(T\) at this point; instead, heat is being spent to change the percentage of ice vs liquid water in the mixture. So indeed, you find lots of places on Earth where the temperature of water or ice is almost exactly 0 °C.

However, in that case of water, I actually gave you a mechanism which explains why the probability density for \(T\) gets a delta-function-like peak at \(T=0\) °C. Could there be a similar "melting-like" mechanism if we replaced \(T\) by the energy density \(\rho\)? And if there is no such mechanism (and let's admit, it seems hard to gradually change \(\rho\)), isn't it a problem for the MPP? Shouldn't a similar mechanism or "qualitative change" of the dynamics be a condition if we claim that some regions of the parameter space are much more likely than others?

There are many interesting questions and "incomplete answers" I can give you but I would like to hear some complete answers and I am not sure that these answers are included – explicitly or implicitly – in this very interesting paper.

by Luboš Motl ( at October 27, 2014 03:25 PM

Tommaso Dorigo - Scientificblogging

New CMS Results
The Large Hadron Collider at the CERN laboratories in Geneva is currently in shutdown, finalizing the upgrades that will allow it to restart next year at the  centre-of-mass energy of 13 TeV - over 60% more than the last 8 TeV run. ATLAS and CMS have collected no more proton-proton collisions since almost two years ago; yet the collaborations are as busy as ever producing physics results from the analysis of the 2012 data.

Rather than focusing on any single result, below I give some highlights of the most recent publications by CMS. Another post will discuss ATLAS results in a few days.

read more

by Tommaso Dorigo at October 27, 2014 01:50 PM

Symmetrybreaking - Fermilab/SLAC

TEDx goes to CERN

Inventors, educators and scientists inspired audiences at this year’s TEDxCERN.

It’s not every day you see entrepreneurs, engineers and particle physicists breaking it down to an electronica remix of recordings from CERN’s air conditioning units.

But then again, TEDxCERN only happens once a year.

TEDxCERN is an independently organized TED-like event hosted by CERN that invites scientists and innovators to share their insight. The goal of this year’s TEDxCERN was to showcase the connections between research and society across multiple disciplines, cultures and continents.

“Science really has a big impact in society, even if it takes a long time for it to be applied and make a difference in your life,” says Claudia Marcelloni, the organizer and curator of TEDxCERN. “We wanted to have a conversation about the biggest problems and where the solutions might be.”

The talks addressed topics such as health, genetically modified organisms, nuclear power and climate change.

“The future depends on science, and if we are going to make the right decisions, both at the personal level and at a global level, we need to be able to think rationally about science,” says James Gillies, CERN’s head of communication and the head of the speakers selection committee for TEDxCERN. “For a big public-facing organization like CERN, I think it is almost a moral obligation for us to do events like TEDxCERN and get the word out about other areas of research.”

The TEDxCERN event also wove information about CERN’s fundamental research into the program. CERN cosmologist Julien Lesgourgues discussed the composition of the universe. And animations produced by the TED-Ed education team explained research into antimatter, cosmic rays and cloud formation.

Marcelloni and Gillies were both pleased with how the event turned out, though Marcelloni admits she was originally a bit nervous about the last-minute addition of electronica DJ Tim Exile to the program.

“I had no idea how it would end, and it was a big risk for this kind of audience, but I was very happy with how it ended,” Marcelloni says. “The dance was not planned.”

The videos of the talks and performances are now available online on the TEDxCERN website.


Like what you see? Sign up for a free subscription to symmetry!

by Sarah Charley at October 27, 2014 01:00 PM

October 26, 2014

Michael Schmitt - Collider Blog

The CMS kinematic edge

Does CMS observe an excess that corresponds to a signal for Supersymmetry? Opinions differ.

This week a paper appeared with the statement “…CMS has reported an intriguing excess of events with respect to the ones expected in the SM” (Huang & Wagner, arXiv:1410.4998). And last month another phenomenology paper appeared with the title “Interpreting a CMS lljjpTmiss Excess With the Golden Cascade of the MSSM” (Allanach, Raklev and Kvellestad arXiv:1409.3532). Both studies are based on a preliminary CMS report (CMS PAS SUS-12-019, Aug. 24 2014), which ends with the statement “We do not observe evidence for a statistically significant signal.”

What is going on here?

The CMS search examines the di-lepton invariant mass distribution for a set of events which have, in addition to two energetic and isolated leptons, missing transverse momentum (pTmiss) and two jets. In cascade decays of SUSY particles, χ02 → χ01 l+l-, a kind of hard edge appears at the phase space limit for the two leptons l+l-, as pointed out many years ago (PRD 55 (1997) 5520). When distinguishing a signal from background, a sharp edge is almost as good as a peak, so this is a nice way to isolate a signal if one exists. An edge can also be observed in decays of sleptons. The CMS search is meant to be as general as possible.

In order to remove the bulk of SM events producing a pair of leptons, a significant missing energy is required, as expected from a pair of neutralinos χ01. Furthermore, other activity in the event is expected (recall that there will be two supersymmetric particles in the event) so the search demands at least two jets. Hence: ll+jj+pTmiss.

A crucial feature of the analysis is motivated by the phenomenology of SUSY cascade decays: for the signal, the two leptons will have the same flavor (ee or μμ), while most of the SM backgrounds will be flavor blind (so eμ also is expected). By comparing the Mll spectrum for same-flavor and for opposite-flavor leptons, an excess might be observed with little reliance on simulations. Only the Drell-Yan background does not appear in the opposite-flavor sample at the same level as in the same-flavor sample, but a hard cut on pTmiss (also called ETmiss) removes most of the DY background. (I am glossing over careful and robust measurements of the relative e and μ reconstruction and trigger efficiencies – see the note for those details, especially Section 4.)

The CMS analyzers make an important distinction between “central” leptons with |η|<1.4 and "forward" leptons 1.6<|η|<2.4 motivated by the idea that supersymmetric particles will be centrally produced due to their high mass, and an excess may be more pronounced when both leptons are central.

A search for a kinematic edge proceed just as you would expect – a series of fits is performed with the edge placed at different points across a wide range of invariant mass Mll. The model for the Mll spectrum has three components: the flavor-symmetric backgrounds, dominated by tt, the DY background and a hypothetical signal. Both the flavor-symmetric and the DY components are described by heuristic analytical functions with several free parameters. The signal is a triangle convolved with a Gaussian to represent the resolution on Mll. Most of the model parameters are determined in fits with enhanced DY contributions, and through the simultaneous fit to the opposite-flavor sample. For the search, only three parameters are free: the signal yield in the central and forward regions and the position of the kinematic edge.

The best fitted value for the edge is y = 78.7±1.4 GeV. At that point, an excess is observed with a local statistical significance of 2.4σ, in the central region. There is no excess in the forward region. Here is the plot:

Fit to the same-flavor dilepton mass distribution

The green triangle represents the fitted signal. The red peak is, of course, the Z resonance. Here is the distribution for the forward region:

Fit to the opposite invariant mass distribution for the forward sample

Fit to the opposite (electron-muon) invariant mass distribution

Comparing the two distributions and ignoring the Z peak, there does indeed seem to be an excess of ee and μμ pairs for Mll < 80 GeV or so. One can understand why the theorists would find this intriguing…

CMS made a second treatment of their data by defining a mass region 20 < Mll < 70 GeV and simply counting the number of events, thereby avoiding any assumptions about the shape of a signal. For this treatment, one wants to compare the data to the prediction, with suitable corrections for efficiencies, etc. Here are the plots:

Comparison of Mll distributions for data (dots) and the prediction (solid lines).  Left: central.  Right: forward.

By eye one can notice a tendency of the real data (dots) to fall above the prediction (solid line histogram). This tendency is much stronger for the events with two central leptons compared to the events with at least one forward lepton. Counting, CMS reports 860 observed compared to 730±40 predicted (central) and 163 observed for 157±16 forward. The significance is 2.6σ for the central di-leptons.

CMS provides a kind of teaser plot, in which they simulate three signals from the production of sbottom squarks. As you can see here, two of the models describe the apparent excess well:

Comparison to sbottom signals

Comparison to sbottom signals

So why is this not a discovery?

First of all, the statistical significance is under 3σ so formally speaking, this not even “evidence.” More importantly, the “look-elsewhere effect” has not been taken into account, as stated clearly in the CMS note. In other words, the significance for the fit is 2.4σ when you choose 78.7 GeV for the position of the edge. If you allow for any position of the edge within some wide range of Mll, then the chance that you observe an excess somewhere in that range is much greater than 1%. Similarly, the counting excess is 2.6σ for the specific box 20 < Mll < 70 GeV, but if you consider many different boxes, the chance to observe such a single excess somewhere is not so small. For this reason, the CMS Collaboration states that there is no statistically significant excess.

That said, the agreement of the simulated sbottom signals with the data is striking, even granted that there are many free parameters here that one can tune to get a good description. The Allanach et al. paper reports a scan in MSSM parameter space to find which physical SUSY parameters are consistent with the CMS data. They impose a few theoretical constraints that are not crucial at this stage, and perform a scan with simulated signals to see which ranges of parameters reproduce the CMS data and also respect constraints from other ATLAS and CMS searches. An example model is shown here:

Example spectrum from Allanach et al.

Example spectrum from Allanach et al.

Wide ranges of sparticle masses are allowed, but there are strong constraints among them coming from the well-measured position of the edge. Constraints from (g-1)μ are respected and the relic density is good. Best of all, prospects for discovering one of these models at Run 2 are good — if such a signal really is present of course.

The Huang & Wagner paper focuses on the sbottom scenario mentioned in the CMS paper, and does a more detailed and refined analysis. They define two similar scenarios, here is the scheme for the first one:

Huang & Wagner sbottom scenario

Huang & Wagner sbottom scenario

They do not perform a scan of parameter space; rather, they select model parameters by hand to provide pertinent examples. They specifically focus on the relic density and (g-2)μ to make sure they their model can explain these facts. They explain their reasoning clearly in their paper. Their hand-tuned model does a nice job matching the CMS data. Of course, it also evades direct searches for sbottoms by both ATLAS and CMS.

What about the ATLAS 8 TeV data? For now, we must wait.

by Michael Schmitt at October 26, 2014 11:29 PM

ZapperZ - Physics and Physicists

Quantum Foam
More educational video on something which you may have heard, but haven't quite understood.


by ZapperZ ( at October 26, 2014 06:01 PM

Peter Coles - In the Dark

Thursday (and Friday and Saturday) Night Fever

It’s been a very strange weekend. Some horrible bug hit me with a fever on Thursday evening, so I had to cancel my Friday appointments.

Then I remembered I was supposed to be in Cardiff this weekend so dragged myself out of bed and onto the train. Although I wasn’t at all comfortable, I slept for most of the journey. I hope I didn’t infect too many other passengers…

An early night on Friday and most of Saturday in bed seem to have helped, though I was clearly still delirious this afternoon when I dreamt that Newcastle managed to beat Spurs 2-1 at White Hart Lane. I also seem to have won another dictionary in the Sunday Independent Crossword competition.

Now the fever seems to have morphed into a fairly standard cold-type thing and I’ll have to face getting an early morning train back to Brighton tomorrow. Hopefully I’ll get back about lunchtime.

by telescoper at October 26, 2014 05:48 PM

October 25, 2014

Geraint Lewis - Cosmic Horizons

The Redshift Drift
Things are crazily busy, with me finishing teaching this week. Some of you may know, that I am writing a book, which is progressing, but more slowly than I hoped. Up to just over 60,000 words, with a goal of about 80 to 90 thousand, so more than half way through.

I know that I have to catch up with papers, and I have another article in The Conversation brewing, but I thought I would write about something interesting. The problem is that my limited brain has been occupied by so many other things that my clear thinking time has been reduced to snippets here and there.

But one thing that has been on my mind is tests of cosmology. Nothing I post here will be new, but you might not know about it. But here goes.

So, the universe is expanding. But how do we know? I've written a little about this previously, but we know that almost 100 years ago, Edwin Hubble discovered his "law", that galaxies are moving away from us, and the further away they are, the faster they are moving. There's a nice article here describing the importance of this, and we end up with a picture that looks something like this
Distance is actually the hard thing to measure, and there are several books that detail astronomers on-off love affair with measuring distances. But how about velocities?

These are measured using the redshift. It's such a simple measurement. In our laboratory, we might see emission from an element, such as hydrogen, at one wavelength, but when we observe it in a distant galaxy, we see it at another, longer, wavelength. The light has been redshifted due to the expansion of the universe (although exactly what this means can be the source for considerable confuddlement).

Here's an illustration of this;
Relating the redshift to a Doppler shift we can turn it into a velocity. As we know, the Hubble law is  what we expect if we use Einstein's theory of relativity to describe the universe. Excellent stuff all around!

One thing we do know is that the expansion rate of the universe is not uniform in time. It was very fast at the Big Bang, slowed down for much of cosmic history, before accelerating due to the presence of dark energy.

So, there we have an interesting question. Due to the expansion of the universe, will the redshift I measure for a galaxy today be the same when I measure it again tomorrow.

This question was asked before I was born and then again several times afterwards. For those that love mathematics, and who doesn't, you get a change of redshift with time that looks like this

(taken from this great paper) where z is the redshift, Ho is Hubble's constant today, while H(z) is Hubble's constant at the time the light was emitted from the galaxy your observing. 

The cool thing is that last term depends upon the energy content of the universe, just how much mass there is, how much radiation, how much dark energy, and all the other cool things that we would like to know, like if dark energy is evolving and and changing, or interacting with matter and radiation. It would be a cool cosmological probe.

Ah, there is a problem! We know that Hubble's constant is about Ho = 72 km/s/Mpc, which seems like a nice sort of number. But if you look closely, you can see that it actually had units of 1/time. So, expressing it in years, this number is about 0.0000000001 per year. This is a small number. Bottom.

But this does not mean that astronomers pack up their bags and head home. No, you look for solutions and see if you can come up with technologies to allow you to measure this tiny shift. I could write an entire post on this, but people are developing laser combs to give extremely accurate measurement of the wavelength in spectra, and actually measure the changing expansion of the Universe in real time!

Why am I writing about this? Because these direct tests of cosmology have always fascinated me, and every so often I start doodling with the cosmological equations to see if I can come up with another one. Often this ends up with a page of squiggles and goes no where, but some times I have what I thing is a new insight.

And this gives me a chance to spruik an older paper of mine, with then PhD student, Madhura Killedar. I still love this stuff!

The evolution of the expansion rate of the Universe results in a drift in the redshift of distant sources over time. A measurement of this drift would provide us with a direct probe of expansion history. The Lyman alpha forest has been recognized as the best candidate for this experiment, but the signal would be weak and it will take next generation large telescopes coupled with ultra-stable high resolution spectrographs to reach the cm/s resolution required. One source of noise that has not yet been assessed is the transverse motion of Lyman alpha absorbers, which varies the gravitational potential in the line of sight and subsequently shifts the positions of background absorption lines. We examine the relationship between the pure cosmic signal and the observed redshift drift in the presence of moving Lyman alpha clouds, particularly the collapsed structures associated with Lyman limit systems (LLSs) and damped Lyman alpha systems (DLAs). Surprisingly, the peculiar velocities and peculiar accelerations both enter the expression, although the acceleration term stands alone as an absolute error, whilst the velocity term appears as a fractional noise component. An estimate of the magnitude of the noise reassures us that the motion of the Lyman alpha absorbers will not pose a threat to the detection of the signal.

by Cusp ( at October 25, 2014 09:53 PM

Geraint Lewis - Cosmic Horizons

Catching the Conversation
Wow!!! Where has time gone! I must apologise for the sluggishness of posts on this blog. I promise you that it is not dead, I have been consumed with a number of other things and not all of it fun. I will get back to interesting posts as soon as possible.

So, here's a couple of articles I've written in the meantime, appearing in The Conversation

One on some of my own research: Dark matter and the Milky Way: more little than large

And the other on proof (or lack of it) in science: Where’s the proof in science? There is none

There's more to come :)

by Cusp ( at October 25, 2014 08:54 PM

Jaques Distler - Musings


Wow! After a decade, Wikipedia finally rolls out MathML rendering. Currently, only available (as an optional preference) to registered users. Hopefully, in a few more years, they’ll make it the default.

Some implementation details are available at Frédéric’s blog.

by distler ( at October 25, 2014 06:19 AM

October 24, 2014

arXiv blog

Why Quantum "Clippers" Will Distribute Entanglement Across The Oceans

The best way to build a global quantum internet will use containerships to carry qubits across the oceans, say physicists.

One possible future for communication is to create a quantum version of the internet that will have the ability, among other things, to send information with perfect security. This network will use entangled photons to transmit information from one locations to another without it passing through the space in between, hence the security.

October 24, 2014 08:36 PM

The Great Beyond - Nature blog

WHO plans for millions of doses of Ebola vaccine by 2015

Posted on behalf of Declan Butler.

The World Health Organization (WHO) announced plans on 24 October to produce millions of doses of two experimental Ebola vaccines by the end of 2015.

The Ebola virus has caused about 5,000 deaths in West Africa during the current epidemic.

The Ebola virus has caused about 5,000 deaths in West Africa during the current epidemic.


Hundreds of thousands of doses should be available to help affected countries before the end of June, the WHO said at the conclusion of a meeting in Geneva, Switzerland. Vaccine makers, high-level government representatives and regulatory and other bodies gathered to discuss the design and timing of planned clinical trials, as well as issues of supply and funding for mass vaccination programmes.

Phase I trials of two vaccine candidates have started, and as many as five other vaccines could begin testing by 2015, says Marie-Paul Kieny, WHO assistant director-general for health systems and innovation.

As of 19 October, Ebola had infected almost 10,000 people in Sierra Leone, Liberia and Guinea and killed around 5,000 of them, the WHO estimates. The true figures are probably higher, as many cases go unreported. With no end to the epidemic yet in sight, a working vaccine could be a game changer.

First clinical trials under way

The two vaccines whose production will be increased are already in early-stage testing in healthy volunteers. One is a chimpanzee adenovirus vaccine containing a surface Ebola protein (ChAd3), developed by the US National Institute of Allergy and Infectious Diseases and drug giant GlaxoSmithKline (GSK). It is being tested in the United States, the United Kingdom and Mali.

The other is a recombinant vesicular stomatitis virus (rVSV) vaccine, developed by the Public Health Agency of Canada and licensed to NewLink Genetics in Ames, Iowa. It is being tested in the United States, with plans to start trials soon in Europe and Africa.

These phase I trials will assess the vaccines’ safety and whether they elicit levels of immune response that have been shown to confer protection in non-human primates. The trials will also assess the dose needed to generate sufficient immune response, which in turn helps to determine how quickly manufacturers can produce doses.

A third candidate is a two-vaccine regimen: one developed by US pharmaceutical company Johnson and Johnson and the US National Institute of Allergy and Infectious Diseases, and another by Bavarian Nordic, a biotechnology company based in Denmark. It will begin phase I testing in the United States and Europe in January. Johnson and Johnson announced on 22 October that it would spend up to US$200 million to fast track the vaccine’s development; it plans to produce more than 1 million doses in 2015, with 250,000 available by May.

Advanced testing

The first phase II and III trials, to test efficacy as well as safety, are set to start in Liberia in December and in Sierra Leone in January. The current plan is to test both the GSK and NewLink vaccines simultaneously, but that could change depending on the results of the ongoing phase I trials. Data from the phase II and III tests are expected by April, Kieny says.

The ‘three-arm’ Liberia trial would test and compare the safety and effectiveness of the two vaccines against each other and a placebo. Each vaccine would be tested on 10,000 subjects, with an equal number of subjects given placebo. This allows researchers to obtain quick, reliable data on how well the vaccines work.

A ‘stepped-wedge’ randomized trial in Sierra Leone would give subjects vaccine sequentially, with no group given a placebo. This is useful for testing products that are expected to benefit patients, and products that are in short supply.

No trial design has yet been fixed for Guinea, where a lack of infrastructure has precluded early testing. If the Liberia and Sierra Leone trials show that the vaccines works and is safe, subsequent trials in Guinea would be used to answer follow-up questions.

Ethical and practical considerations

The Sierra Leone trial will enrol at least 8,000 health-care workers, and other frontline responders, such as ambulance drivers and burial workers. The Liberia trial might include health-care workers, but these would not be the primary study population, Kieny says.

Any decision to give a placebo to health-care and other frontline workers will be controversial; many consider it to be unethical, given these individuals’ work caring for Ebola patients, and the risks that they face in doing so.

Mass vaccinations are usually only carried out after years of trials to accumulate full safety and efficacy data. The proposed timeline for Ebola vaccine development is therefore unprecedented.

If existing public-health interventions to control Ebola outbreaks begin to slow the epidemic, the need for mass vaccination will lessen, Kieny says. But if the epidemic continues to expand, the WHO could consider expanding vaccination programmes.

In the meantime, the WHO and its partners are considering how best to engage with communities to prepare for vaccination programmes. Another issue is simply determining how to keep vaccine at –80 degrees Celsius, the temperature needed to maintain its efficacy. This will require specialized refrigerators and the establishment of cold supply chains to affected areas.

Also to be determined is who will pay for mass vaccination. Kieny says simply that “money will not be an issue”. Aid groups and governments have begun to pledge support for such efforts. Médecins Sans Frontières (MSF, also known as Doctors Without Borders) has said that it will create a fund for Ebola vaccination, and the European Union has committed €200 million (US$255 million). The Gavi vaccine alliance, the main sponsor of routine vaccinations in low-income countries, is also looking at how it could bring its vast resources and experience to the table. It will put a plan to its board in December as to what role it could have in any Ebola mass vaccination.

by Lauren Morello at October 24, 2014 07:47 PM

astrobites - astro-ph reader's digest

Mind the Gaps

“Music is the silence between the notes.” – Claude Debussy

Astronomical data gathered over time has gaps. For instance, when using a ground-based telescope, there is the pesky fact that roughly half of every 24-hour period is lit by the Sun. Or, the star you want to look at isn’t above the horizon, or clouds are blocking it. Even the most reliable space telescopes suffer from occasional pauses in their otherwise constant watchfulness.

Why are gaps a problem? Can’t astronomers just analyze the short chunks of data that don’t have gaps? Besides, no observation is truly continuous: there is always some gap between data points. Why should slightly longer or shorter gaps really make a difference?

The answer: Fourier transforms.

The Fourier transform “is like a mathematical prism—you feed in a wave and it spits out the ingredients of that wave.” (Read more of the superb Nautilus piece explaining the Fourier transform here.) It is an incredibly versatile data analysis tool. But in order for it to work perfectly, there are a couple important rules. First, the starting wave, or dataset, can have no beginning or end. Second, all the data points must be evenly spaced.

Of course, those of us leftward-of-mathematician on the field purity scale know that gap-free, infinite observations are never going to happen. So we need to fill gaps and mask edges. Today’s paper takes a look at how this is often done (spoiler: not carefully enough), and proposes a new gap-filling method to better preserve all the information in stellar light curves.


Periodogram (calculated using a Fast Fourier Transform) of a Delta Scuti star’s pulsations. The blue version has gaps in the light curve filled with simple linear interpolation, while the red version has used Pascual-Granada et al.’s new MIARMA algorithm to fill the gaps. Figure 1 from the paper.

The image above compares two slightly different Fourier transforms of a pulsating Delta Scuti star light curve, observed by the CoRoT satellite. The blue transform uses a common gap-filling technique: linear interpolation. This is simply drawing a straight line from the last point before a gap to the first point after it and pretending points on that line are observations with the same regular spacing as the real data. In contrast, the red transform uses a new algorithm called MIARMA to fill gaps in the light curve. As you can see, the frequencies and their heights and patterns are very different between these two methods. Since the main goal of asteroseismology is to learn about the insides of stars by studying their oscillation frequencies, you had better be sure you are studying the correct frequencies!

Pascual-Granada et al. create the MIARMA algorithm using an autoregressive moving average (ARMA) model. In essence, it looks at data on either side of a gap to predict what happens after the gap ends and before it begins—an autoregression, and it does this many times for each gap with different combinations of data points—a moving average.


Filling gaps in a solar-type star’s light curve spanning two days. Blue points are CoRoT observations, red points are gaps filled with MIARMA, and green points are gaps filled with linear interpolation. Figure 6 from the paper.

To demonstrate MIARMA preserves information better than linear interpolation, the authors test it on three different variable stars observed with CoRoT. They study the Delta Scuti pulsator described above, a Be star with longer time variations, and a rapidly-varying solar-type star.

Overall, MIARMA makes the biggest difference for the two stars with light curves that vary more slowly. For these, frequency spikes present in the linear interpolation case match with how often gaps tend to occur. The MIARMA Fourier transform lacks these telltale spikes and is free of aliasing—a common problem in signal processing in which incorrect frequencies and amplitudes are inferred because you aren’t recording data often enough. But the choice of gap-filler does not matter as much for more rapidly-varying solar-type stars. This makes sense because the typical separation between two gaps is long compared to how quickly the star is varying. As a result, the scientifically interesting frequencies are less susceptible to being affected by the gaps.

The authors report that their new method will be used to process all CoRoT data going forward, and can be adapted to work with Kepler data too. This is an important reminder that scientists must deeply understand their data. Sometimes the most problematic data points are none at all.

by Meredith Rawls at October 24, 2014 06:59 PM

The Great Beyond - Nature blog

US research ethics agency upholds decision on informed consent

United States regulators are standing by their decision that parents were not properly informed of the risks of a clinical trial in which premature babies received different levels of oxygen supplementation.

From 2005 to 2009, the Surfactant, Positive Pressure, and Oxygenation Randomized Trial (SUPPORT) trial randomly assigned 1,316 premature babies to receive one of two levels of oxygen supplementation in an effort to test which level was best. Even though the lower level was associated with increased risk of brain damage and possibly death, and the higher level with blindness, the study leaders said that they did not disclose these risks to parents because all ranges of oxygen used in the trial were considered to be within the medically appropriate range at the time.

The study was supported by the US National Institutes of Health (NIH). On 7 March 2013, the US Office of Human Research Protections (OHRP) issued a letter determining that the trial investigators had not adequately informed parents about the risks to their babies in the SUPPORT trial. The NIH and many researchers disputed the decision, arguing that it would impede “comparative effectiveness research” studies that are designed to test the best use of approved interventions. Parents of children in the trial, however, and others supported the OHRP’s determination that parents hadn’t received adequate information. The two sides clashed at a meeting convened by the NIH and the OHRP in August 2013.

Today, 24 October 2014, the OHRP has issued guidance reiterating and clarifying its position on what types of risks must be disclosed to study subjects in comparative effectiveness research studies such as SUPPORT. The agency has determined that risks of the intervention must be disclosed to study participants even if the risks are considered acceptable according to current medical guidelines, if the study intends to evaluate these risks and if the patients’ risks will change when they enrol in the study.

The OHRP said that even though both the low and high levels of oxygen supplementation were considered within the acceptable range, “the key issue is that the treatment and possible risks infants were exposed to in the research were different from the treatment and possible risks they would have been exposed to if they had not been in the trial”.

“[F]or the great majority of infants in the trial, it is likely that their participation altered the level of oxygen they received compared to what they would have received had they not participated,” the OHRP added.

The agency said further that if a trial is designed to compare the risks of potential side effects of a treatment already in use, then the risks are “reasonably foreseeable” and that prospective study participants should be made aware of it.

“If a specific risk has been identified as significant enough that it is important for the Federal government to spend taxpayer money to better understand the extent or nature of that risk, then that risk is one that prospective subjects should be made aware of so that they can decide if they want to be exposed to it,” OHRP said.

The guidance is open to comments until 24 December.

by Erika Check Hayden at October 24, 2014 05:32 PM

The Great Beyond - Nature blog

Western Australia abandons shark cull

Western Australia Premier Colin Barnett

Government of Western Australia

The state of Western Australia is abandoning a controversial shark-culling programme, but has also gained the right to deploy deadly baited lines for animals that pose an “imminent threat”.

The programme, run by the state government off several Western Australian beaches, had been heavily criticized by scientists when it was announced in 2013. It was due to run until 2017, and had caught at least 170 sharks using hooks suspended from drums moored to the sea floor.

In September the state’s own Environmental Protection Agency halted it. State Premier Colin Barnett then applied to the national government for permission to resume it, but today he announced that his government had ended that effort. “We have withdrawn the application after reaching agreement with the Commonwealth which enables us to take immediate action when there is an imminent threat,” said Barnett.

Under an agreement with the national government, Western Australia will be able to kill sharks in future to deal with a shark that has attacked or with one that it thinks poses a threat. Protocols for how this would happen are now in development.

This apparent concession from the national government has drawn some concern from those celebrating the end of the cull.

“I remain concerned that drum lines could be used in some instances as part of emergency measures and particularly that this could occur without Federal approval,” said Rachel Siewert, the marine spokeswoman for the Australian Greens, in a statement.

The Western Australia cull is also drawing renewed attention to the longstanding cull in Queensland, which continues unabated.

by Daniel Cressey at October 24, 2014 02:47 PM

Symmetrybreaking - Fermilab/SLAC

Cosmic inflation

Cosmic inflation refers to a period of rapid, accelerated expansion that scientists think took place about 14 billion years ago.

Cosmic inflation refers to a period of rapid, accelerated expansion that scientists think took place about 14 billion years ago.

Our universe has likely never grown as quickly as it did during that period. Faster than the blink of an eye, the whole universe expanded so that an area the size of an atom was suddenly the size of a grapefruit.

Scientists think this expansion was driven by the potential energy of the inflaton field, a new field that turned on just after the big bang.

Support for the theory of cosmic inflation comes from the Cosmic Microwave Background, or CMB, a pattern of light released when the early universe first cooled enough for particles to travel freely through it.

Although nearly uniform, the CMB contains ripples. Scientists think these were caused by tiny quantum fluctuations that were amplified to huge scales by cosmic inflation.

Scientists study cosmic inflation through experiments at telescopes, such as the Planck satellite and BICEP2 at the South Pole. These experiments measure elements of the CMB, looking for the footprints of inflation.

When inflation ended, the expansion of our universe began to slow down. But then another influence took over, pushing it back to an accelerating rate. This influence is thought to be dark energy.


Like what you see? Sign up for a free subscription to symmetry!

by Rhianna Wisniewski at October 24, 2014 01:00 PM

Quantum Diaries

Where the wind goes sweeping ’round the ring?

I travel a lot for my work in particle physics, but it’s usually the same places over and over again — Fermilab, CERN, sometimes Washington to meet with our gracious supporters from the funding agencies.  It’s much more interesting to go someplace new, and especially somewhere that has some science going on that isn’t particle physics.  I always find myself trying to make connections between other people’s work and mine.

This week I went to a meeting of the Council of the Open Science Grid that was hosted by the Oklahoma University Supercomputing Center for Education and Research in Norman, OK.  It was already interesting that I got to visit Oklahoma, where I had never been before.  (I think I’m up to 37 states now.)  But we held our meeting in the building that hosts the National Weather Center, which gave me an opportunity to take a tour of the center and learn a bit more about how research in meteorology and weather forecasting is done.

OU is the home of the largest meteorology department in the country, and the center hosts a forecast office of the National Weather Service (which produces forecasts for central and western Oklahoma and northern Texas, at the granularity of one hour and one kilometer) and the National Severe Storms Laboratory (which generates storm watches and warnings for the entire country — I saw the actual desk where the decisions get made!).  So how is the science of the weather like and not like the science that we do at the LHC?

(In what follows, I offer my sincere apologies to meteorologists in case I misinterpreted what I learned on my tour!)

Both are fields that can generate significant amounts of data that need to be interpreted to obtain a scientific result.  As has been discussed many times on the blog, each LHC experiment records petabytes of data each year.  Meteorology research is performed by much smaller teams of observers, which makes it hard to estimate their total data volume, but the graduate student who led our tour told us that he is studying a mere three weather events, but he has more than a terabyte of data to contend with — small compared to what a student on the LHC might have to handle, but still significant.

But where the two fields differ is what limits the rate at which the data can be understood.  At the LHC, it’s all about the processing power needed to reconstruct the raw data by performing the algorithms that turn the voltages read out from millions of amplifiers into the energies and momenta of individual elementary particles.  We know what the algorithms for this are, we know how to code them; we just have to run them a lot.  In meteorology, the challenge is getting to the point where you can even make the data interpretable in a scientific sense.  Things like radar readings still need to be massaged by humans to become sensible.  It is a very labor-intensive process, akin to the work done by the “scanner girls” of the particle physics days of yore, who carefully studied film emulsions by eye to identify particle tracks.  I do wonder what the prospects are in meteorology for automating this process so that it can be handed off to humans instead.  (Clearly this has to apply more towards forefront research in the field about how tornadoes form and the like, rather than to the daily weather predictions that just tell you the likelihood of tornado-forming conditions.)

Weather forecasting data is generally public information, accessible by anyone.  The National Weather Service publishes it in a form that has already had some processing done on it so that it can be straightforwardly ingested by others.  Indeed, there is a significant private weather-forecasting industry that makes use of this, and sells products with value added to the NWS data.  (For instance, you could buy a forecast much more granular than that provided by the NWS, e.g. for the weather at your house in ten-minute intervals.)  Many of these companies rent space in buildings within a block of the National Weather Center.  The field of particle physics is still struggling with how to make our data publicly available (which puts us well behind many astronomy projects which make all of their data public within a few years of the original observations).  There are concerns about how to provide the data in a form that will allow people who are not experts to learn something from the data without making mistakes.  But there has been quite a lot of progress in this in recent years, especially as it has been recognized that each particle physics experiment creates a unique dataset that will probably never be replicated in the future.  We can expect an increasing number of public data releases in the next few years.  (On that note, let me point out the NSF-funded Data and Software Preservation for Open Science (DASPOS) project that I am associated with on its very outer edges, which is working on some aspects of the problem.)  However, I’d be surprised if anyone starts up a company that will sell new interpretations of LHC data!

Finally, here’s one thing that the weather and the LHC has in common — they’re both always on!  Or, at least we try to run the LHC for every minute possible when the accelerator is operational.  (Remember, we are currently down for upgrades and will start up again this coming spring.)  The LHC experiments have physicists on on duty 24 hours a day, monitoring data quality and ready to make repairs to the detectors should they be needed.  Weather forecasters are also on shift at the forecasting center and the severe-storm center around the clock.  They are busy looking at data being gathered by their own instruments, but also from other sources.  For instance, when there are reports of tornadoes near Oklahoma City, the local TV news stations often send helicopters out to go take a look.  The forecasters watch the TV news to get additional perspectives on the storm.

Now, if only the weather forecasters on shift could make repairs to the weather just like our shifters can fix the detector!

by Ken Bloom at October 24, 2014 05:14 AM

October 23, 2014

Clifford V. Johnson - Asymptotia

Reading Storm…
Screen Shot 2014-10-23 at 09.55.07For a while back there earlier this week I was in a storm of reading duties of the sort that I hope not to see again in a while. A lot of it had to be put off at the end of the week before because I wanted to prepare my talk for Sunday, which took a little more time than I'd planned since I wanted to do some drawings for it. All of it had a deadline. Monday was to see me participating in a podcast at the USC Bedrosian Center to discuss the book "Beyond the University: Why Liberal Education Matters", by Michael S. Roth. I had the book for about six weeks, and started reading it when I first got it... but found that I was getting through it too fast too early and wanted to have it fresher in my mind for the podcast, so I held off until closer to the date. Unfortunately, this then clashed with two promotion dossiers that got scheduled for a Tuesday meeting, both from book-heavy fields, and so that added three books on language, representation, business and history (tangled up in a fascinating way) that I can't tell you about since the proceedings of the relevant committee are confidential. Then I remembered that a Ph.D. thesis exam had been moved from the previous week to that same Tuesday (and I had put off the reading) and so I had a thesis to read as well. (Not to mention all the dossier letters, statements, committee reports, and so forth that come from reading two promotion dossiers...) A lot of the reading is also fun, but it's certainly hard work and one is reading while taking careful notes for later reference, in a lot of the instances. I always end up surprising myself with how much fun I have learning about topics far beyond my field when I read promotions dossiers for other areas. I'm certainly not an expert (and that is not why I'm called into service in such cases) so I'm reading with an eye on seeing what the quality of scholarship is, and what the voice of the person writing is like. These are things that (if you are not of the tedious point of view that your own field of inquiry is somehow king of the disciplines (a view we physicists all too often seem to have)) can be glimpsed and sometimes firmly perceived by wading deep into the pool of their work and keeping an open mind. I strongly recommend the Roth book about what the point [...] Click to continue reading this post

by Clifford at October 23, 2014 05:06 PM

The Great Beyond - Nature blog

Fundamental overhaul of China’s competitive funding

On 20 October, the Chinese government announced the passage of a reform plan that will fundamentally reshape research in the country.

By 2017, the main competitive government funding initiatives will be eliminated. This includes the ’863′ and ’973′ programmes, two channels for large grants that have been at the heart of modern China’s development of science and technology infrastructure since being established in 1986 and 1997, respectively.

Xi Jinping, General Secretary of the Communist Party of China, is behind reforms to overhaul research in the country.

Xi Jinping, General Secretary of the Communist Party of China, is behind reforms to overhaul research in the country.

By Antilong (Own work) [CC-BY-SA-3.0], via Wikimedia Commons

The government announcement noted that wastefulness and fragmented management has led to overlaps and inefficient use of funds for science and technology, and the need for a unified platform for distributing grants. As new funding programmes have been added over the years, competitive funding has become divided among some 100 competitive schemes overseen by about 30 different governmental departments.

Although efforts to reorganize science in China are already underway,  the latest reform will be comprehensive. Science and technology spending by the central government was 77.4 billion yuan renminbi (US$12.6 billion) in 2006 but jumped to 236 billion yuan renminbi in 2013, 11.6% of the central government’s direct public expenditure. Some 60% of this is competitive funding, and subject to change under under the new reforms. To maintain stability, the overhaul will not affect the remaining 40%, which covers operation costs for research institutes and key state laboratories.

The new plan, jointly drafted by the ministries of science and technology and the ministry of finance, will reorganize competitive funding into five new channels: the National Natural Science Foundation (which currently distributes many of the small-scale competitive grants); national science and technology major projects; key national research and development programmes; a special fund to guide technological innovation; and special projects for developing human resources and infrastructure. These five will be managed under a new science and technology agency that will unify planning and assessment of scientific projects.




by David Cyranoski at October 23, 2014 02:57 PM

Symmetrybreaking - Fermilab/SLAC

Australia’s first dark matter experiment

A proposed dark matter experiment would use two underground detectors, one in each hemisphere.

Physicists are hoping to hit pay dirt with a proposed experiment—the first of its kind in the Southern Hemisphere—that would search for traces of dark matter more than a half mile below ground in Victoria, Australia.

The current plan, now being explored by an international team, is for two new, identical dark matter experiments to be installed and operated in parallel—one at an underground site at Grand Sasso National Laboratory in Italy, and the other at the Stawell Gold Mine in Australia.

“An experiment of this significance could ultimately lead to the discovery of dark matter,” says Elisabetta Barberio of the ARC Centre of Excellence for Particle Physics at the Terascale (CoEPP) and the University of Melbourne, who is Australian project leader for the proposed experiment.

The experiment proposal was discussed during a two-day workshop on dark matter in September. Work could begin on the project as soon as 2015 if it gathers enough support. “We’re looking at logistics and funding sources,” Barberio says.

The experiments would be modeled after the DAMA experiment at Gran Sasso, now called DAMA/LIBRA, which in 1998 found a possible sign of dark matter.

DAMA/LIBRA looks for seasonal modulation, an ebb and flow in the amount of potential dark matter signals it sees depending on the time of year.

If the Milky Way is surrounded by a halo of dark matter particles, then the sun is constantly moving through it, as is the Earth. The Earth’s rotation around the sun causes the two to spend half of the year moving in the same direction and the other half moving in opposite directions. During the six months in which the Earth and sun are cooperating, a dark matter detector on the Earth will move faster through the dark matter particles, giving it more opportunities to catch them.

This seasonal difference appears in the data from DAMA/LIBRA, but no other experiment has been able to confirm this as a sign of dark matter.

For one thing, the changes in the signal could be caused on other factors that change by the season.

“There are environmental effects—different characteristics of the atmosphere—in winter and summer that are clearly reversed if you go from the Northern to the Southern hemisphere,” says Antonio Masiero, vice president for the Italian National Institute of Nuclear Physics (INFN) and a member of the Italian delegation collaborating on the proposal, which also includes Gran Sasso Director Stefano Ragazzi. If the results matched up at both sites at the same time of year, that would help to rule out such effects.

The Australian mine hosting the proposed experiment could also house scientific experiments from different fields.

“It wouldn’t be limited to particle physics and could include experiments involving biology, geosciences and engineering,” Barberio says. “These could include neutrino detection, nuclear astrophysics, geothermal energy extraction and carbon sequestration, and subsurface imaging and sensing.”

Preliminary testing has begun at the mine site down to depths of about 880 meters, about 200 meters above the proposed experimental site. Regular mining operations are scheduled to cease at Stawell in the next few years.

The ARC Centre of Excellence for All-Sky Astrophysics (CAASTRO), the local government in the Victoria area, and the mine operators have joined forces with COEPP and INFN to support the proposal.


Like what you see? Sign up for a free subscription to symmetry!

by Glenn Roberts Jr. at October 23, 2014 02:56 PM

astrobites - astro-ph reader's digest

Today’s Partial Solar Eclipse

Fig. 1: Comparing lunar and solar eclipses. (Credit: Encyclopedia Britannica, Inc.)

This is a special post for our North American readers. In case you haven’t heard, there will be a partial solar eclipse today, visible from Mexico, Canada and the USA. “But wait!” you might say, “Didn’t we just have an eclipse?!” Yes, in fact, there was a lovely total lunar eclipse on October 8th. But today’s eclipse is a rarer solar eclipse.

Lunar vs. Solar Eclipses

Remember that a lunar eclipse occurs when the Earth passes between the Sun and Moon, casting a shadow on the Moon. Everyone on the half of the Earth that’s facing the Moon can view a lunar eclipse. But a solar eclipse occurs when the Moon passes between the Earth and Sun, casting a much smaller shadow on the Earth. You can only view a solar eclipse if you’re standing on the part of the Earth where the shadow falls, so it’s rarer to see a solar eclipse than a lunar eclipse (unless you’re willing to go on an eclipse expedition).

Lucky Coincidences in the Earth-Moon-Sun System

We get to enjoy solar eclipses thanks to a happy coincidence in the Earth-Moon-Sun system. The Moon is much smaller than the Sun, of course, but it’s just the right distance from Earth that the Moon and Sun have roughly the same apparent size in the sky. This means that when the Moon passes between us and the Sun, it’s just the right size to block the Sun’s light during a total solar eclipse.

This is lucky for eclipse fans, and if the Earth, Moon, and Sun orbited on the same plane, we’d get to see a total solar eclipse about once a month. In fact, we’d probably bored with them after a while. However, the tilt of the Moon’s orbit keeps eclipses from occurring more often than about every 18 months. Fig. 2 shows why the Moon’s 5° tilt makes solar eclipses rare.

Fig. 2: Diagram showing the effect of the tilt of the Moon’s orbit on the frequency of solar eclipses. (Credit:

Observing the Eclipse

Today’s solar eclipse is only a partial eclipse, meaning that the Moon does not pass directly between the Earth and Sun, so it will only appear to take a bite out of the Sun rather than blocking it completely. This means that if you’re not trying to observe the eclipse, you probably won’t even notice it, since the Sun’s light will not dim significantly. So, what’s the best way to observe the eclipse?

Images of an Eclipse on the ground

Fig. 3: The gaps between the leaves of a tree can act as pinhole cameras, projecting the image of the eclipse onto the ground. (Credit: CSIRO)

First and most importantly, practice safe eclipse-viewing. Looking directly at the Sun for a long period can cause permanent damage to your vision, even during a partial eclipse. Looking through a telescope or binoculars at the Sun is even more dangerous! Even using a telescope to project an image of the eclipse onto a piece of paper can be dangerous, because the heat collected inside the telescope can damage it or break glass lenses or eyepieces.

Your best option is to visit a local observatory or astronomy club and take advantage of their expertise and equipment. If you’re on your own, you could try to get your hands on some eclipse glasses. Sunglasses won’t protect your eyes, even if they’re UV-blocking sunglasses, and neither will looking through undeveloped film or tinted glass. You need some solar-rated filtered glasses. I picked up a pair during the Venus transit back in 2012.

If it’s too late to find some filtered glasses, you can easily create a pinhole camera to project the image of the eclipse onto the ground or a piece of paper. If you want to build a pinhole camera, there are some great instructions here, but you can also just use anything that allows light to pass through one or more small holes, like a sieve or even the leaves of a tree (like in Fig. 3).

If it’s too cloudy where you are, you can check out one of the live streams from the Coca-Cola Science Center or the Mt. Lemmon SkyCenter. Fig. 4 shows a map of the start times of the eclipse, and the amount of light that will be blocked by the Moon, for different regions in North America. Sky & Telescope also has a table listing eclipse start and end times for different cities.

Visibility of October 23, 2014, solar eclipse

Fig. 4: Visibility map of the October 23 partial solar eclipse. (Credit: Sky & Telescope / Leah Tiscione)

If you do have access to a properly-filtered telescope, be sure to look for the huge sunspot, AR 2192. At 100,000 km across, this sunspot is big enough to swallow the Earth, and just yesterday it spit out an X-class solar flare. AR 2192 is easily visible through a solar telescope.

Even if you don’t have any equipment, I hope you North Americans will take the opportunity to step outside with a last-minute pinhole camera today to catch the eclipse. The next solar eclipse visible from North America won’t be until 2017, when we’ll get the chance to see a rare total solar eclipse!

by Erika Nesvold at October 23, 2014 12:59 PM

The n-Category Cafe

Why It Matters

One interesting feature of the Category Theory conference in Cambridge last month was that lots of the other participants started conversations with me about the whole-population, suspicionless surveillance that several governments are now operating. All but one were enthusiastically supportive of the work I’ve been doing to try to get the mathematical community to take responsibility for its part in this, and I appreciated that very much.

The remaining one was a friend who wasn’t unsupportive, but said to me something like “I think I probably agree with you, but I’m not sure. I don’t see why it matters. Persuade me!”

Here’s what I replied.

“A lot of people know now that the intelligence agencies are keeping records of almost all their communications, but they can’t bring themselves to get worked up about it. And in a way, they might be right. If you, personally, keep your head down, if you never do anything that upsets anyone in power, it’s unlikely that your records will end up being used against you.

But that’s a really self-centred attitude. What about people who don’t keep their heads down? What about protesters, campaigners, activists, people who challenge the establishment — people who exercise their full democratic rights? Freedom from harassment shouldn’t depend on you being a quiet little citizen.

“There’s a long history of intelligence agencies using their powers to disrupt legitimate activism. The FBI recorded some of Martin Luther King’s extramarital liaisons and sent the tape to his family home, accompanied by a letter attempting to blackmail him into suicide. And there have been many many examples since then (see below).

“Here’s the kind of situation that worries me today. In the UK, there’s a lot of debate at the moment about the oil extraction technique known as fracking. The government has just given permission for the oil industry to use it, and environmental groups have been protesting vigorously.

“I don’t have strong opinions on fracking myself, but I do think people should be free to organize and protest against it without state harassment. In fact, the state should be supporting people in the exercise of their democratic rights. But actually, any anti-fracking group would be sensible to assume that it’s the object of covert surveillance, and that the police are working against it, perhaps by employing infiltrators — because they’ve been doing that to other environmental groups for years.

“It’s the easiest thing in the world for politicians to portray anti-fracking activists as a danger to the UK’s economic well-being, as a threat to national energy security. That’s virtually terrorism! And once someone’s been labelled with the T word, it immediately becomes trivial to justify using all that surveillance data that the intelligence agencies routinely gather. And I’m not exaggerating — anti-terrorism laws really have been used against environmental campaigners in the recent past.

“Or think about gay rights. Less than fifty years ago, sex between men in England was illegal. This law was enforced, and it ruined people’s lives. For instance, my academic great-grandfather Alan Turing was arrested under this law and punished with chemical castration. He’s widely thought to have killed himself as a direct result. But today, two men in England can not only have sex legally, they can marry with the full endorsement of the state.

“How did this change so fast? Not by people writing polite letters to the Times, or by going through official parliamentary channels (at least, not only by those means). It was mainly through decades of tough, sometimes dangerous, agitation, campaigning and protest, by small groups and by courageous individual citizens.

“By definition, anyone campaigning for anything to be decriminalized is siding with criminals against the establishment. It’s the easiest thing in the world for politicians to portray campaigners like this as a menace to society, a grave threat to law and order. Any nation state with the ability to monitor, infiltrate, harass and disrupt such “menaces” will be very sorely tempted to use it. And again, that’s no exaggeration: in the US at least, this has happened to gay rights campaigners over and over again, from the 1950s to nearly the present day, even sometimes — ludicrously — in the name of fighting terrorism (1, 2, 3, 4).

“So government surveillance should matter to you in a very direct way if you’re involved in any kind of activism or advocacy or protest or campaigning or dissent. It should also matter to you if you’re not, but you quietly support any of this activism — or if you reap its benefits. Even if you don’t (which is unlikely), it matters if you simply want to live in a society where people can engage in peaceful activism without facing disruption or harassment by the state. And it matters more now than it ever did before, because government surveillance powers are orders of magnitude greater than they’ve ever been before.”

That’s roughly what I said. I think we then talked a bit about mathematicians’ role in enabling whole-population surveillance. Here’s Thomas Hales’s take on this:

If privacy disappears from the face of the Earth, mathematicians will be some of the primary culprits.

Of course, there are lots of other reasons why the activities of the NSA, GCHQ and their partners might matter to you. Maybe you object to industrial espionage being carried out in the name of national security, or the NSA supplying data to the CIA’s drone assassination programme (“we track ‘em, you whack ‘em”), or the raw content of communications between Americans being passed en masse to Israel, or the NSA hacking purely civilian infrastructure in China, or government agencies intercepting lawyer-client and journalist-source communications, or that the existence of mass surveillance leads inevitably to self-censorship. Or maybe you simply object to being watched, for the same reason you close the bathroom door: you’re not doing anything to be ashamed of, you just want some privacy. But the activism point is the one that resonates most deeply with me personally, and it seemed to resonate with my friend too.

You may think I’m exaggerating or scaremongering — that the enormous power wielded by the US and UK intelligence agencies (among others) could theoretically be used against legitimate citizen activism, but hasn’t been so far.

There’s certainly an abstract argument against this: it’s simply human nature that if you have a given surveillance power available to you, and the motive to use it, and the means to use it without it being known that you’ve done so, then you very likely will. Even if (for some reason) you believe that those currently wielding these powers have superhuman powers of self-restraint, there’s no guarantee that those who wield them in future will be equally saintly.

But much more importantly, there’s copious historical evidence that governments routinely use whatever surveillance powers they possess against whoever they see as troublemakers, even if this breaks the law. Without great effort, I found 50 examples in the US and UK alone — read on.

Six overviews

If you’re going to read just one thing on government surveillance of activists, I suggest you make it this:

Among many other interesting points, it reminds us that this isn’t only about “leftist” activism — three of the plaintiffs in this case are pro-gun organizations.

Here are some other good overviews:

And here’s a short but incisive comment from journalist Murtaza Hussain.

50 episodes of government surveillance of activists

Disclaimer   Journalism about the activities of highly secretive organizations is, by its nature, very difficult. Even obtaining the basic facts can be a major feat. Obviously, I can’t attest to the accuracy of all these articles — and the entries in the list below are summaries of the articles linked to, not claims I’m making myself. As ever, whether you believe what you read is a judgement you’ll have to make for yourself.


1. FBI surveillance of War Resisters League (1, 2), continuing in 2010 (1)


2. FBI surveillance of the National Association for the Advancement of Colored People (1)

3. FBI “surveillance program against homosexuals” (1)


4. FBI’s Sex Deviate programme (1)

5. FBI’s Cointelpro projects aimed at “surveying, infiltrating, discrediting, and disrupting domestic political organizations”, and NSA’s Project Minaret targeted leading critics of Vietnam war, including senators, civil rights leaders and journalists (1)

6. FBI attempted to blackmail Martin Luther King into suicide with surveillance tape (1)

7. NSA intercepted communications of antiwar activists, including Jane Fonda and Dr Benjamin Spock (1)

8. Harassment of California student movement (including Stephen Smale’s free speech advocacy) by FBI, with support of Ronald Reagan (1, 2)


9. FBI surveillance and attempted deportation of John Lennon (1)

10. FBI burgled the office of the psychiatrist of Pentagon Papers whistleblower Daniel Ellsberg (1)


11. Margaret Thatcher had the Canadian national intelligence agency CSEC surveil two of her own ministers (1, 2, 3)

12. MI5 tapped phone of founder of Women for World Disarmament (1)

13. Ronald Reagan had the NSA tap the phone of congressman Michael Barnes, who opposed Reagan’s Central America policy (1)


14. NSA surveillance of Greenpeace (1)

15. UK police’s “undercover work against political activists” and “subversives”, including future home secretary Jack Straw (1)

16. UK undercover policeman Peter Francis “undermined the campaign of a family who wanted justice over the death of a boxing instructor who was struck on the head by a police baton” (1)

17. UK undercover police secretly gathered intelligence on 18 grieving families fighting to get justice from police (1, 2)

18. UK undercover police spied on lawyer for family of murdered black teenager Stephen Lawrence; police also secretly recorded friend of Lawrence and his lawyer (1, 2)

19. UK undercover police spied on human rights lawyers Bindmans (1)

20. GCHQ accused of spying on Scottish trade unions (1)


21. US military spied on gay rights groups opposing “don’t ask, don’t tell” (1)

22. Maryland State Police monitored nonviolent gay rights groups as terrorist threat (1)

23. NSA monitored email of American citizen Faisal Gill, including while he was running as Republican candidate for Virginia House of Delegates (1)

24. NSA surveillance of Rutgers professor Hooshang Amirahmadi and ex-California State professor Agha Saeed (1)

25. NSA tapped attorney-client conversations of American lawyer Asim Ghafoor (1)

26. NSA spied on American citizen Nihad Awad, executive director of the Council on American-Islamic Relations, the USA’s largest Muslim civil rights organization (1)

27. NSA analyst read personal email account of Bill Clinton (date unknown) (1)

28. Pentagon counterintelligence unit CIFA monitored peaceful antiwar activists (1)

29. Green party peer and London assembly member Jenny Jones was monitored and put on secret police database of “domestic extremists” (1, 2)

30. MI5 and UK police bugged member of parliament Sadiq Khan (1, 2)

31. Food Not Bombs (volunteer movement giving out free food and protesting against war and poverty) labelled as terrorist group and infiltrated by FBI (1, 2, 3)

32. Undercover London police infiltrated green activist groups (1)

33. Scottish police infiltrated climate change activist organizations, including anti-airport expansion group Plane Stupid (1)

34. UK undercover police had children with activists in groups they had infiltrated (1)

35. FBI infiltrated Muslim communities and pushed those with objections to terrorism (and often mental health problems) to commit terrorist acts (1, 2, 3)


36. California gun owners’ group Calguns complains of chilling effect of NSA surveillance on members’ activities (1, 2, 3)

37. GCHQ and NSA surveilled Unicef and head of Economic Community of West African States (1)

38. NSA spying on Amnesty International and Human Rights Watch (1)

39. CIA hacked into computers of Senate Intelligence Committee, whose job it is to oversee the CIA
(1, 2, 3, 4, 5, 6; bonus: watch CIA director John Brennan lie that it didn’t happen, months before apologizing)

40. CIA obtained legally protected, confidential email between whistleblower officials and members of congress, regarding CIA torture programme (1)

41. Investigation suggests that CIA “operates an email surveillance program targeting senate intelligence staffers” (1)

42. FBI raided homes and offices of Anti-War Committee and Freedom Road Socialist Organization, targeting solidarity activists working with Colombians and Palestinians (1)

43. Nearly half of US government’s terrorist watchlist consists of people with no recognized terrorist group affiliation (1)

44. FBI taught counterterrorism agents that mainstream Muslims are “violent” and “radical”, and used presentations about the “inherently violent nature of Islam” (1, 2, 3)

45. GCHQ has developed tools to manipulate online discourse and activism, including changing outcomes of online polls, censoring videos, and mounting distributed denial of service attacks (1, 2)

46. Green member of parliament Caroline Lucas complains that GCHQ is intercepting her communications (1)

47. GCHQ collected IP addresses of visitors to Wikileaks websites (1, 2)

48. The NSA tracks web searches related to privacy software such as Tor, as well as visitors to the website of the Linux Journal (calling it an “extremist forum”) (1, 2, 3)

49. UK police attempt to infiltrate anti-racism, anti-fascist and environmental groups, anti-tax-avoidance group UK Uncut, and politically active Cambridge University students (1, 2)

50. NSA surveillance impedes work of investigative journalists and lawyers (1, 2, 3, 4, 5).

Back to mathematics

As mathematicians, we spend much of our time studying objects that don’t exist anywhere in the world (perfect circles and so on). But we exist in the world. So, being a mathematician sometimes involves addressing real-world concerns.

For instance, Vancouver mathematician Izabella Laba has for years been writing thought-provoking posts on sexism in mathematics. That’s not mathematics, but it’s a problem that implicates every mathematician. On this blog, John Baez has written extensively on the exploitative practices of certain publishers of mathematics journals, the damage it does to the universities we work in, and what we can do about it.

I make no apology for bringing political considerations onto a mathematical blog. The NSA is a huge employer of mathematicians — over 1000 of us, it claims. Like it or not, it is part of our mathematical environment. Both the American Mathematical Society and London Mathematical Society are now regularly publishing articles on the role of mathematicians in enabling government surveillance, in recognition of our responsibility for it. As a recent New York Times article put it:

To say mathematics is political is not to diminish it, but rather to recognize its greater meaning, promise and responsibilities.

by leinster ( at October 23, 2014 07:51 AM

The n-Category Cafe

Where Do Probability Measures Come From?

Guest post by Tom Avery

Tom (here Tom means me, not him — Tom) has written several times about a piece of categorical machinery that, when given an appropriate input, churns out some well-known mathematical concepts. This machine is the process of constructing the codensity monad of a functor.

In this post, I’ll give another example of a well-known concept that arises as a codensity monad; namely probability measures. This is something that I’ve just written a paper about.

The Giry monads

Write <semantics>Meas<annotation encoding="application/x-tex">\mathbf{Meas}</annotation></semantics> for the category of measurable spaces (sets equipped with a <semantics>σ<annotation encoding="application/x-tex">\sigma</annotation></semantics>-algebra of subsets) and measurable maps. I’ll also write <semantics>I<annotation encoding="application/x-tex">I</annotation></semantics> for the unit interval <semantics>[0,1]<annotation encoding="application/x-tex">[0,1]</annotation></semantics>, equipped with the Borel <semantics>σ<annotation encoding="application/x-tex">\sigma</annotation></semantics>-algebra.

Let <semantics>ΩMeas<annotation encoding="application/x-tex">\Omega \in \mathbf{Meas}</annotation></semantics>. There are lots of different probability measures we can put on <semantics>Ω<annotation encoding="application/x-tex">\Omega</annotation></semantics>; write <semantics>GΩ<annotation encoding="application/x-tex">G\Omega</annotation></semantics> for the set of all of them.

Is <semantics>GΩ<annotation encoding="application/x-tex">G\Omega</annotation></semantics> a measurable space? Yes: An element of <semantics>GΩ<annotation encoding="application/x-tex">G\Omega</annotation></semantics> is a function that sends measurable subsets of <semantics>Ω<annotation encoding="application/x-tex">\Omega</annotation></semantics> to numbers in <semantics>I<annotation encoding="application/x-tex">I</annotation></semantics>. Turning this around, we have, for each measurable <semantics>AΩ<annotation encoding="application/x-tex">A \subseteq \Omega</annotation></semantics>, an evaluation map <semantics>ev A:GΩI<annotation encoding="application/x-tex">ev_A \colon G\Omega \to I</annotation></semantics>. Let’s give <semantics>GΩ<annotation encoding="application/x-tex">G\Omega</annotation></semantics> the smallest <semantics>σ<annotation encoding="application/x-tex">\sigma</annotation></semantics>-algebra such that all of these are measurable.

Is <semantics>G<annotation encoding="application/x-tex">G</annotation></semantics> a functor? Yes: Given a measurable map <semantics>g:ΩΩ<annotation encoding="application/x-tex">g \colon \Omega \to \Omega'</annotation></semantics> and <semantics>πGΩ<annotation encoding="application/x-tex">\pi \in G\Omega</annotation></semantics>, we can define the pushforward <semantics>Gg(π)<annotation encoding="application/x-tex">G g(\pi)</annotation></semantics> of <semantics>π<annotation encoding="application/x-tex">\pi</annotation></semantics> along <semantics>g<annotation encoding="application/x-tex">g</annotation></semantics> by

<semantics>Gg(π)(A)=π(g 1A)<annotation encoding="application/x-tex"> G g(\pi)(A') = \pi(g^{-1} A') </annotation></semantics>

for measurable <semantics>AΩ<annotation encoding="application/x-tex">A' \subseteq \Omega'</annotation></semantics>.

Is <semantics>G<annotation encoding="application/x-tex">G</annotation></semantics> a monad? Yes: Given <semantics>ωΩ<annotation encoding="application/x-tex">\omega \in \Omega</annotation></semantics> we can define <semantics>η(ω)GΩ<annotation encoding="application/x-tex">\eta(\omega) \in G\Omega</annotation></semantics> by

<semantics>η(ω)(A)=χ A(ω)<annotation encoding="application/x-tex"> \eta(\omega)(A) = \chi_A (\omega) </annotation></semantics>

where <semantics>A<annotation encoding="application/x-tex">A</annotation></semantics> is a measurable subset of <semantics>Ω<annotation encoding="application/x-tex">\Omega</annotation></semantics> and <semantics>χ A<annotation encoding="application/x-tex">\chi_A</annotation></semantics> is its characteristic function. In other words <semantics>η(ω)<annotation encoding="application/x-tex">\eta(\omega)</annotation></semantics> is the Dirac measure at <semantics>ω<annotation encoding="application/x-tex">\omega</annotation></semantics>. Given <semantics>ρGGΩ<annotation encoding="application/x-tex">\rho \in G G\Omega</annotation></semantics>, let

<semantics>μ(ρ)(A)= GΩev Adρ<annotation encoding="application/x-tex"> \mu(\rho)(A) = \int_{\G\Omega} ev_A \,\mathrm{d}\rho </annotation></semantics>

for measurable <semantics>AΩ<annotation encoding="application/x-tex">A \subseteq \Omega</annotation></semantics>, where <semantics>ev A:GΩI<annotation encoding="application/x-tex">\ev_A \colon G\Omega \to I</annotation></semantics> is as above.

This is the Giry monad <semantics>𝔾=(G,η,μ)<annotation encoding="application/x-tex">\mathbb{G} = (G,\eta,\mu)</annotation></semantics>, first defined (unsurprisingly) by Giry in “A categorical approach to probability theory”.

A finitely additive probability measure <semantics>π<annotation encoding="application/x-tex">\pi</annotation></semantics> is just like a probability measure, except that it is only well-behaved with respect to finite disjoint unions, rather than arbitrary countable disjoint unions. More precisely, rather than having

<semantics>π( i=1 A i)= i=1 π(A i)<annotation encoding="application/x-tex"> \pi\left(\bigcup_{i=1}^{\infty} A_i\right) = \sum_{i=1}^{\infty} \pi(A_i) </annotation></semantics>

for disjoint <semantics>A i<annotation encoding="application/x-tex">A_i</annotation></semantics>, we just have

<semantics>π( i=1 nA i)= i=1 nπ(A i)<annotation encoding="application/x-tex"> \pi\left(\bigcup_{i=1}^{n} A_i\right) = \sum_{i=1}^{n} \pi(A_i) </annotation></semantics>

for disjoint <semantics>A i<annotation encoding="application/x-tex">A_i</annotation></semantics>.

We could repeat the definition of the Giry monad with “probability measure” replaced by “finitely additive probability measure”; doing so would give the finitely additive Giry monad <semantics>𝔽=(F,η,μ)<annotation encoding="application/x-tex">\mathbb{F} = (F,\eta,\mu)</annotation></semantics>. Every probability measure is a finitely additive probability measure, but not all finitely additive probability measures are probability measures. So <semantics>𝔾<annotation encoding="application/x-tex">\mathbb{G}</annotation></semantics> is a proper submonad of <semantics>𝔽<annotation encoding="application/x-tex">\mathbb{F}</annotation></semantics>.

The Kleisli category of <semantics>𝔾<annotation encoding="application/x-tex">\mathbb{G}</annotation></semantics> is quite interesting. Its objects are just the measurable spaces, and the morphisms are a kind of non-deterministic map called a Markov kernel or conditional probability distribution. As a special case, a discrete space equipped with an endomorphism in the Kleisli category is a discrete-time Markov chain.

I’ll explain how the Giry monads arise as codensity monads, but first I’d like to mention a connection with another example of a codensity monad; namely the ultrafilter monad.

An ultrafilter <semantics>𝒰<annotation encoding="application/x-tex">\mathcal{U}</annotation></semantics> on a set <semantics>X<annotation encoding="application/x-tex">X</annotation></semantics> is a set of subsets of <semantics>X<annotation encoding="application/x-tex">X</annotation></semantics> satisfying some properties. So <semantics>𝒰<annotation encoding="application/x-tex">\mathcal{U}</annotation></semantics> is a subset of the powerset <semantics>𝒫X<annotation encoding="application/x-tex">\mathcal{P}X</annotation></semantics> of <semantics>X<annotation encoding="application/x-tex">X</annotation></semantics>, and is therefore determined by its characteristic function, which takes values in <semantics>{0,1}I<annotation encoding="application/x-tex">\{0,1\} \subseteq I</annotation></semantics>. In other words, an ultrafilter on <semantics>X<annotation encoding="application/x-tex">X</annotation></semantics> can be thought of as a special function

<semantics>𝒫XI.<annotation encoding="application/x-tex"> \mathcal{P}X \to I. </annotation></semantics>

It turns out that “special function” here means “finitely additive probability measure defined on all of <semantics>𝒫X<annotation encoding="application/x-tex">\mathcal{P}X</annotation></semantics> and taking values in <semantics>{0,1}<annotation encoding="application/x-tex">\{0,1\}</annotation></semantics>”.

So the ultrafilter monad on <semantics>Set<annotation encoding="application/x-tex">\mathbf{Set}</annotation></semantics> (which sends a set to the set of ultrafilters on it) is a primitive version of the finitely additive Giry monad. With this in mind, and given the fact that the ultrafilter monad is the codensity monad of the inclusion of the category of finite sets into the category of sets, it is not that surprising that the Giry monads are also codensity monads. In particular, we might expect <semantics>𝔽<annotation encoding="application/x-tex">\mathbb{F}</annotation></semantics> to be the codensity monad of some functor involving spaces that are “finite” in some sense, and for <semantics>𝔾<annotation encoding="application/x-tex">\mathbb{G}</annotation></semantics> we’ll need to include some information pertaining to countable additivity.

Integration operators

If you have a measure on a space then you can integrate functions on that space. The converse is also true: if you have a way of integrating functions on a space then you can extract a measure.

There are various ways of making this precise, the most famous of which is the Riesz-Markov-Kakutani Representation Theorem:

Theorem. Let <semantics>X<annotation encoding="application/x-tex">X</annotation></semantics> be a compact Hausdorff space. Then the space of finite, signed Borel measures on <semantics>X<annotation encoding="application/x-tex">X</annotation></semantics> is canonically isomorphic to

<semantics>NVS(Top(X,),)<annotation encoding="application/x-tex"> \mathbf{NVS}(\mathbf{Top}(X,\mathbb{R}),\mathbb{R}) </annotation></semantics>

as a normed vector space, where <semantics>Top<annotation encoding="application/x-tex">\mathbf{Top}</annotation></semantics> is the category of topological spaces, and <semantics>NVS<annotation encoding="application/x-tex">\mathbf{NVS}</annotation></semantics> is the category of normed vector spaces.

Given a finite, signed Borel measure <semantics>π<annotation encoding="application/x-tex">\pi</annotation></semantics> on <semantics>X<annotation encoding="application/x-tex">X</annotation></semantics>, the corresponding map <semantics>Top(X,)<annotation encoding="application/x-tex">\mathbf{Top}(X,\mathbb{R}) \to \mathbb{R}</annotation></semantics> sends a function to its integral with respect to <semantics>π<annotation encoding="application/x-tex">\pi</annotation></semantics>. There are various different versions of this theorem that go by the same name.

My paper contains the following more modest version, which is a correction of a claim by Sturtz.

Proposition. Finitely additive probability measures on a measurable space <semantics>Ω<annotation encoding="application/x-tex">\Omega</annotation></semantics> are canonically in bijection with functions <semantics>ϕ:Meas(Ω,I)I<annotation encoding="application/x-tex">\phi \colon \mathbf{Meas}(\Omega,I) \to I</annotation></semantics> that are

  • affine: if <semantics>f,gMeas(Ω,I)<annotation encoding="application/x-tex">f,g \in \mathbf{Meas}(\Omega,I)</annotation></semantics> and <semantics>rI<annotation encoding="application/x-tex">r \in I</annotation></semantics> then

<semantics>ϕ(rf+(1r)g)=rϕ(f)+(1r)ϕ(g),<annotation encoding="application/x-tex"> \phi(r f + (1-r)g) = r\phi(f) + (1-r)\phi(g), </annotation></semantics>


  • weakly averaging: if <semantics>r¯<annotation encoding="application/x-tex">\bar{r}</annotation></semantics> denotes the constant function with value <semantics>r<annotation encoding="application/x-tex">r</annotation></semantics> then <semantics>ϕ(r¯)=r<annotation encoding="application/x-tex">\phi(\bar{r}) = r</annotation></semantics>.

Call such a function a finitely additive integration operator. The bijection restricts to a correspondence between (countably additive) probability measures and functions <semantics>ϕ<annotation encoding="application/x-tex">\phi</annotation></semantics> that additionally

  • respect limits: if <semantics>f nMeas(Ω,I)<annotation encoding="application/x-tex">f_n \in \mathbf{Meas}(\Omega,I)</annotation></semantics> is a sequence of functions converging pointwise to <semantics>0<annotation encoding="application/x-tex">0</annotation></semantics> then <semantics>ϕ(f n)<annotation encoding="application/x-tex">\phi(f_n)</annotation></semantics> converges to <semantics>0<annotation encoding="application/x-tex">0</annotation></semantics>.

Call such a function an integration operator. The integration operator corresponding to a probability measure <semantics>π<annotation encoding="application/x-tex">\pi</annotation></semantics> sends a function <semantics>f<annotation encoding="application/x-tex">f</annotation></semantics> to

<semantics> Ωfdπ,<annotation encoding="application/x-tex"> \int_{\Omega}f \mathrm{d}\pi, </annotation></semantics>

which justifies the name. In the other direction, given an integration operator <semantics>ϕ<annotation encoding="application/x-tex">\phi</annotation></semantics>, the value of the corresponding probability measure on a measurable set <semantics>AΩ<annotation encoding="application/x-tex">A \subseteq \Omega</annotation></semantics> is <semantics>ϕ(χ A)<annotation encoding="application/x-tex">\phi(\chi_A)</annotation></semantics>.

These bijections are measurable (with respect to a natural <semantics>σ<annotation encoding="application/x-tex">\sigma</annotation></semantics>-algebra on the set of finitely additive integration operators) and natural in <semantics>Ω<annotation encoding="application/x-tex">\Omega</annotation></semantics>, so they define isomorphisms of endofunctors of <semantics>Meas<annotation encoding="application/x-tex">\mathbf{Meas}</annotation></semantics>. Hence we can transfer the monad structures across the isomorphisms, and obtain descriptions of the Giry monads in terms of integration operators.

The Giry monads via codensity monads

So far so good. But what does this have to do with codensity monads? First let’s recall the definition of a codensity monad. I won’t go into a great deal of detail; for more information see Tom’s first post on the topic.

Let <semantics>U:<annotation encoding="application/x-tex">U \colon \mathbb{C} \to \mathcal{M}</annotation></semantics> be a functor. The codensity monad of <semantics>U<annotation encoding="application/x-tex">U</annotation></semantics> is the right Kan extension of <semantics>U<annotation encoding="application/x-tex">U</annotation></semantics> along itself. This consists of a functor <semantics>T U:<annotation encoding="application/x-tex">T^U \colon \mathcal{M} \to \mathcal{M}</annotation></semantics> satisfying a universal property, which equips <semantics>T U<annotation encoding="application/x-tex">T^U</annotation></semantics> with a canonical monad structure. The codensity monad doesn’t always exist, but it will whenever <semantics><annotation encoding="application/x-tex">\mathbb{C}</annotation></semantics> is small and <semantics><annotation encoding="application/x-tex">\mathcal{M}</annotation></semantics> is complete. You can think of <semantics>T U<annotation encoding="application/x-tex">T^U</annotation></semantics> as a generalisation of the monad induced by the adjunction between <semantics>U<annotation encoding="application/x-tex">U</annotation></semantics> and its left adjoint that makes sense when the left adjoint doesn’t exist. In particular, when the left adjoint does exist, the two monads coincide.

The end formula for right Kan extensions gives

<semantics>T Um= c[(m,Uc),Uc],<annotation encoding="application/x-tex"> T^U m = \int_{c \in \mathbb{C}} [\mathcal{M}(m,U c),U c], </annotation></semantics>

where <semantics>[(m,Uc),Uc]<annotation encoding="application/x-tex">[\mathcal{M}(m,U c),U c]</annotation></semantics> denotes the <semantics>(m,Uc)<annotation encoding="application/x-tex">\mathcal{M}(m,U c)</annotation></semantics> power of <semantics>Uc<annotation encoding="application/x-tex">U c</annotation></semantics> in <semantics><annotation encoding="application/x-tex">\mathcal{M}</annotation></semantics>, i.e. the product of <semantics>(m,Uc)<annotation encoding="application/x-tex">\mathcal{M}(m,U c)</annotation></semantics> (a set) copies of <semantics>Uc<annotation encoding="application/x-tex">U c</annotation></semantics> (an object of <semantics><annotation encoding="application/x-tex">\mathcal{M}</annotation></semantics>) in <semantics><annotation encoding="application/x-tex">\mathcal{M}</annotation></semantics>.

It doesn’t matter too much if you’re not familiar with ends because we can give an explicit description of <semantics>T Um<annotation encoding="application/x-tex">T^U m</annotation></semantics> in the case that <semantics>=Meas<annotation encoding="application/x-tex">\mathcal{M} = \mathbf{Meas}</annotation></semantics>: The elements of <semantics>T UΩ<annotation encoding="application/x-tex">T^U\Omega</annotation></semantics> are families <semantics>α<annotation encoding="application/x-tex">\alpha</annotation></semantics> of functions

<semantics>α c:Meas(Ω,Uc)Uc<annotation encoding="application/x-tex"> \alpha_c \colon \mathbf{Meas}(\Omega, U c) \to U c </annotation></semantics>

that are natural in <semantics>c<annotation encoding="application/x-tex">c \in \mathbb{C}</annotation></semantics>. For each <semantics>c<annotation encoding="application/x-tex">c \in \mathbb{C}</annotation></semantics> and measurable <semantics>f:ΩUc<annotation encoding="application/x-tex">f \colon \Omega \to U c</annotation></semantics> we have <semantics>ev f:T UΩI<annotation encoding="application/x-tex">\ev_f \colon T^U \Omega \to I</annotation></semantics> mapping <semantics>α<annotation encoding="application/x-tex">\alpha</annotation></semantics> to <semantics>α c(f)<annotation encoding="application/x-tex">\alpha_c (f)</annotation></semantics>. The <semantics>σ<annotation encoding="application/x-tex">\sigma</annotation></semantics>-algebra on <semantics>T UΩ<annotation encoding="application/x-tex">T^U \Omega</annotation></semantics> is the smallest such that each of these maps is measurable.

All that’s left is to say what we should choose <semantics><annotation encoding="application/x-tex">\mathbb{C}</annotation></semantics> and <semantics>U<annotation encoding="application/x-tex">U</annotation></semantics> to be in order to get the Giry monads.

A subset <semantics>c<annotation encoding="application/x-tex">c</annotation></semantics> of a real vector space <semantics>V<annotation encoding="application/x-tex">V</annotation></semantics> is convex if for any <semantics>x,yc<annotation encoding="application/x-tex">x,y \in c</annotation></semantics> and <semantics>rI<annotation encoding="application/x-tex">r \in I</annotation></semantics> the convex combination <semantics>rx+(1r)y<annotation encoding="application/x-tex">r x + (1-r)y</annotation></semantics> is also in <semantics>c<annotation encoding="application/x-tex">c</annotation></semantics>, and a map <semantics>h:cc<annotation encoding="application/x-tex">h \colon c \to c'</annotation></semantics> between convex sets is called affine if it preserves convex combinations. So there’s a category of convex sets and affine maps between them. We will be interested in certain full subcategories of this.

Let <semantics>d 0<annotation encoding="application/x-tex">d_0</annotation></semantics> be the (convex) set of sequences in <semantics>I<annotation encoding="application/x-tex">I</annotation></semantics> that converge to <semantics>0<annotation encoding="application/x-tex">0</annotation></semantics> (it is a subset of the vector space <semantics>c 0<annotation encoding="application/x-tex">c_0</annotation></semantics> of all real sequences converging to <semantics>0<annotation encoding="application/x-tex">0</annotation></semantics>). Now we can define the categories of interest:

  • Let <semantics><annotation encoding="application/x-tex">\mathbb{C}</annotation></semantics> be the category whose objects are all finite powers <semantics>I n<annotation encoding="application/x-tex">I^n</annotation></semantics> of <semantics>I<annotation encoding="application/x-tex">I</annotation></semantics>, with all affine maps between them.

  • Let <semantics>𝔻<annotation encoding="application/x-tex">\mathbb{D}</annotation></semantics> be the category whose objects are all finite powers of <semantics>I<annotation encoding="application/x-tex">I</annotation></semantics>, together with <semantics>d 0<annotation encoding="application/x-tex">d_0</annotation></semantics>, and all affine maps between them.

All the objects of <semantics><annotation encoding="application/x-tex">\mathbb{C}</annotation></semantics> and <semantics>𝔻<annotation encoding="application/x-tex">\mathbb{D}</annotation></semantics> can be considered as measurable spaces (as subspaces of powers of <semantics>I<annotation encoding="application/x-tex">I</annotation></semantics>), and all the affine maps between them are then measurable, so we have (faithful but not full) inclusions <semantics>U:Meas<annotation encoding="application/x-tex">U \colon \mathbb{C} \to \mathbf{Meas}</annotation></semantics> and <semantics>V:𝔻Meas<annotation encoding="application/x-tex">V \colon \mathbb{D} \to \mathbf{Meas}</annotation></semantics>.

Theorem. The codensity monad of <semantics>U<annotation encoding="application/x-tex">U</annotation></semantics> is the finitely additive Giry monad, and the codensity monad of <semantics>V<annotation encoding="application/x-tex">V</annotation></semantics> is the Giry monad.

Why should this be true? Let’s start with <semantics>U<annotation encoding="application/x-tex">U</annotation></semantics>. An element of <semantics>T UΩ<annotation encoding="application/x-tex">T^U \Omega</annotation></semantics> is a family of functions

<semantics>α I n:Meas(Ω,I n)I n.<annotation encoding="application/x-tex"> \alpha_{I^n} \colon\mathbf{Meas}(\Omega,I^n) \to I^n. </annotation></semantics>

But a map into <semantics>I n<annotation encoding="application/x-tex">I^n</annotation></semantics> is determined by its composites with the projections to <semantics>I<annotation encoding="application/x-tex">I</annotation></semantics>, and these projections are affine. This means that <semantics>α<annotation encoding="application/x-tex">\alpha</annotation></semantics> is completely determined by <semantics>α I<annotation encoding="application/x-tex">\alpha_{I}</annotation></semantics>, and the other components are obtained by applying <semantics>α I<annotation encoding="application/x-tex">\alpha_{I}</annotation></semantics> separately in each coordinate. In other words, an element of <semantics>T UΩ<annotation encoding="application/x-tex">T^U \Omega</annotation></semantics> is a special sort of function

<semantics>Meas(Ω,I)I.<annotation encoding="application/x-tex"> \mathbf{Meas}(\Omega, I) \to I. </annotation></semantics>

Look familiar? As you might guess, the functions with the above domain and codomain that define elements of <semantics>T UΩ<annotation encoding="application/x-tex">T^U \Omega</annotation></semantics> are precisely the finitely additive integration operators.

The affine and weakly averaging properties of <semantics>α I<annotation encoding="application/x-tex">\alpha_{I}</annotation></semantics> are enforced by naturality with respect to certain affine maps. For example, the naturality square involving the affine map

<semantics>rπ 1+(1r)π 2:I 2I<annotation encoding="application/x-tex"> r\pi_1 + (1-r)\pi_2 \colon I^2 \to I </annotation></semantics>

(where <semantics>π i<annotation encoding="application/x-tex">\pi_i</annotation></semantics> are the projections) forces <semantics>α I<annotation encoding="application/x-tex">\alpha_I</annotation></semantics> to preserve convex combinations of the form <semantics>rf+(1r)g<annotation encoding="application/x-tex">r f + (1-r)g</annotation></semantics>. The weakly averaging condition comes from naturality with respect to constant maps.

How is the situation different for <semantics>T V<annotation encoding="application/x-tex">T^V</annotation></semantics>? As before <semantics>αT VΩ<annotation encoding="application/x-tex">\alpha \in T^V \Omega</annotation></semantics> is determined by <semantics>α I<annotation encoding="application/x-tex">\alpha_I</annotation></semantics>, and <semantics>α d 0<annotation encoding="application/x-tex">\alpha_{d_0}</annotation></semantics> is obtained by applying <semantics>α I<annotation encoding="application/x-tex">\alpha_I</annotation></semantics> in each coordinate, thanks to naturality with respect to the projections. A measurable map <semantics>f:Ωd 0<annotation encoding="application/x-tex">f \colon \Omega \to d_0</annotation></semantics> is a sequence of maps <semantics>f n:ΩI<annotation encoding="application/x-tex">f_n \colon \Omega \to I</annotation></semantics> converging pointwise to <semantics>0<annotation encoding="application/x-tex">0</annotation></semantics>, and

<semantics>α d 0(f)=(α I(f i)) i=1 .<annotation encoding="application/x-tex"> \alpha_{d_0}(f) = (\alpha_I(f_i))_{i=1}^{\infty}. </annotation></semantics>

But <semantics>α d 0(f)d 0<annotation encoding="application/x-tex">\alpha_{d_0}(f) \in d_0</annotation></semantics>, so <semantics>α I(f i)<annotation encoding="application/x-tex">\alpha_I(f_i)</annotation></semantics> must converge to <semantics>0<annotation encoding="application/x-tex">0</annotation></semantics>. So <semantics>α I<annotation encoding="application/x-tex">\alpha_I</annotation></semantics> is an integration operator!

The rest of the proof consists of checking that these assignments <semantics>αα I<annotation encoding="application/x-tex">\alpha \mapsto \alpha_{I}</annotation></semantics> really do define isomorphisms of monads.

It’s natural to wonder how much you can alter the categories <semantics><annotation encoding="application/x-tex">\mathbb{C}</annotation></semantics> and <semantics>𝔻<annotation encoding="application/x-tex">\mathbb{D}</annotation></semantics> without changing the codensity monads. Here’s a result to that effect:

Proposition. The categories <semantics><annotation encoding="application/x-tex">\mathbb{C}</annotation></semantics> and <semantics>𝔻<annotation encoding="application/x-tex">\mathbb{D}</annotation></semantics> can be replaced by the monoids of affine endomorphisms of <semantics>I 2<annotation encoding="application/x-tex">I^2</annotation></semantics> and <semantics>d 0<annotation encoding="application/x-tex">d_0</annotation></semantics> respectively (regarded as 1-object categories, with the evident functors to <semantics>Meas<annotation encoding="application/x-tex">\mathbf{Meas}</annotation></semantics>) without changing the codensity monads.

This gives categories of convex sets that are minimal such that their inclusions into <semantics>Meas<annotation encoding="application/x-tex">\mathbf{Meas}</annotation></semantics> give rise to the Giry monads. Here I mean minimal in the sense that they contain the fewest objects with all affine maps between them. They are not uniquely minimal; there are other convex sets whose monoids of affine endomorphisms also give rise to the Giry monads.

This result gives yet another characterisation of (finitely and countably) additive probability measures: a probability measure on <semantics>Ω<annotation encoding="application/x-tex">\Omega</annotation></semantics> is an <semantics>End(d 0)<annotation encoding="application/x-tex">\mathrm{End}(d_0)</annotation></semantics>-set morphism

<semantics>Meas(Ω,d 0)d 0,<annotation encoding="application/x-tex"> \mathbf{Meas}(\Omega,d_0) \to d_0, </annotation></semantics>

where <semantics>End(d 0)<annotation encoding="application/x-tex">\mathrm{End}(d_0)</annotation></semantics> is the monoid of affine endomorphisms of <semantics>d 0<annotation encoding="application/x-tex">d_0</annotation></semantics>. Similarly for finitely additive probability measures, with <semantics>d 0<annotation encoding="application/x-tex">d_0</annotation></semantics> replaced by <semantics>I 2<annotation encoding="application/x-tex">I^2</annotation></semantics>.

What about maximal categories of convex sets giving rise to the Giry monads? I don’t have a definitive answer to this question, but you can at least throw in all bounded, convex subsets of Euclidean space:

Proposition. Let <semantics><annotation encoding="application/x-tex">\mathbb{C}'</annotation></semantics> be the category of all bounded, convex subsets of <semantics> n<annotation encoding="application/x-tex">\mathbb{R}^n</annotation></semantics> (where <semantics>n<annotation encoding="application/x-tex">n</annotation></semantics> varies) and affine maps. Let <semantics>𝔻<annotation encoding="application/x-tex">\mathbb{D}'</annotation></semantics> be <semantics><annotation encoding="application/x-tex">\mathbb{C}'</annotation></semantics> but with <semantics>d 0<annotation encoding="application/x-tex">d_0</annotation></semantics> adjoined. Then replacing <semantics><annotation encoding="application/x-tex">\mathbb{C}</annotation></semantics> by <semantics><annotation encoding="application/x-tex">\mathbb{C}'</annotation></semantics> and <semantics>𝔻<annotation encoding="application/x-tex">\mathbb{D}</annotation></semantics> by <semantics>𝔻<annotation encoding="application/x-tex">\mathbb{D}'</annotation></semantics> does not change the codensity monads.

The definition of <semantics>𝔻<annotation encoding="application/x-tex">\mathbb{D}'</annotation></semantics> is a bit unsatisfying; <semantics>d 0<annotation encoding="application/x-tex">d_0</annotation></semantics> feels (and literally is) tacked on. It would be nice to have a characterisation of all the subsets of <semantics> <annotation encoding="application/x-tex">\mathbb{R}^{\mathbb{N}}</annotation></semantics> (or indeed all the convex sets) that can be included in <semantics>𝔻<annotation encoding="application/x-tex">\mathbb{D}'</annotation></semantics>. But so far I haven’t found one.

by leinster ( at October 23, 2014 07:50 AM

October 22, 2014

Quantum Diaries

Have we detected Dark Matter Axions?

An interesting headline piqued my interest when browsing the social networking and news website Reddit the other day. It simply said:

“The first direct detection of dark matter particles may have been achieved.”

Well, that was news to me! 
Obviously, the key word here is “may”. Nonetheless, I was intrigued, not being aware of any direct detection experiments publishing such results around this time. As a member of LUX, there are usually collaboration-wide emails sent out when a big paper is published by a rival group, most recently the DarkSide-50 results . Often an email like this is followed by a chain of comments, both good and bad, from the senior members of our group. I can’t imagine there being a day where I think I could read a paper and instantly have intelligent criticisms to share like those guys – but maybe when I’ve been in the dark matter business for 20+ years I will!

It is useful to look at other work similar to our own. We can learn from the mistakes and successes of the other groups within our community, and most of the time rivalry is friendly and professional. 
So obviously I took a look at this claimed direct detection. Note that there are three methods to dark matter detection, see figure. To summarise quickly,

The three routes to dark matter detection

  • Direct detection is the observation of an interaction of a dark matter particle with a standard model one
  • Indirect detection is the observation of annihilation products that have no apparent standard model source and so are assumed to be the products of dark matter annihilation.
  • Production is the measurement of missing energy and momentum in a particle interaction (generally a collider experiment) that could signify the creation of dark matter (this method must be very careful, as this is how the neutrinos are measured in collider experiments).

So I was rather surprised to find the article linked was about a space telescope – the XMM-Newton observatory. These sort of experiments are usually for indirect detection. The replies on the Reddit link reflected my own doubt – aside from the personification of x-rays, this comment was also my first thought:

“If they detected x-rays who are produced by dark matter axions then it’s not direct detection.”

These x-rays supposedly come from a particle called an axion – a dark matter candidate. But to address the comment, I considered LUX, a direct dark matter detector, where what we are actually detecting is photons. These are produced by the recoil of a xenon nuclei that interacted with a dark matter particle, and yet we call it direct – because the dark matter has interacted with a standard model particle, the xenon. 
So to determine whether this possible axion detection is direct, we need to understand the effect producing the x-rays. And for that, we need to know about axions.

I haven’t personally studied axions much at all. At the beginning of my PhD, I read a paper called “Expected Sensitivity to Galactic/Solar Axions and Bosonic Super-WIMPs based on the Axio-electric Effect in Liquid Xenon Dark Matter Detectors” – but I couldn’t tell you a single thing from that paper now, without re-reading it. After some research I have a bit more understanding under my belt, and for those of you that are physicists, I can summarise the idea:

  • The axion is a light boson, proposed by Roberto Peccei and Helen Quinn in 1977 to solve the strong CP problem (why does QCD not break CP-symmetry when there is no theoretical reason it shouldn’t?).
  • The introduction of the particle causes the strong CP violation to go to zero (by some fancy maths that I can’t pretend to understand!).
It has been considered as a cold dark matter candidate because it is neutral and very weakly interacting, and could have been produced with the right abundance.
Conversion of an axion to  a photon within a magnetic field (Yamanaka, Masato et al)

Conversion of an axion to a photon within a magnetic field (Yamanaka, Masato et al)

For non-physicists, the key thing to understand is that the axion is a particle predicted by a separate theory (nothing to do with dark matter) that solves another problem in physics. It just so happens that its properties make it a suitable candidate for dark matter. Sounds good so far – the axion kills two birds with one stone. We could detect a dark matter axion via an effect that converts an axion to an x-ray photon within a magnetic field. The XMM-Newton observatory orbits the Earth and looks for x-rays produced by the conversion of an axion within the Earth’s magnetic field. Although there is no particular interaction with a standard model particle (one is produced), the axion is not annihilating to produce the photons, so I think it is fair to call this direct detection.

What about the actual results? What has actually been detected is a seasonal variation in the cosmic x-ray background. The conversion signal is expected to be greater in summer due to the changing visibility of the magnetic field region facing the sun, and that’s exactly what was observed. In the paper’s conclusion the authors state:

“On the basis of our results from XMM-Newton, it appears plausible that axions – dark matter particle candidates – are indeed produced in the core of the Sun and do indeed convert to soft X-rays in the magnetic field of the Earth, giving rise to a significant, seasonally-variable component of the 2-6 keV CXB”



Conversion of solar axions into photons within the Earth’s magnetic field (University of Leicester)

Note the language used – “it appears plausible”. This attitude of physicists to always be cautious and hold back from bold claims is a wise one – look what happened to BICEP2. It is something I am personally becoming familiar with, last week having come across a lovely LUX event that passed my initial cuts and looked very much like it could have been a WIMP. My project partner from my masters degree at the University of Warwick is now a new PhD student at UCL – and he takes great joy in embarrassing me in whatever way he can. So after I shared my findings with him, he told everyone we came across that I had found WIMPs. Even upon running into my supervisor, he asked “Have you seen Sally’s WIMP?”. I was not pleased – that is not a claim I want to make as a mere second year PhD student. Sadly, but not unexpectedly, my “WIMP” has now been cut away. But not for one second did I truly believe it could have been one – surely there’s no way I‘m going to be the one that discovers dark matter! (Universe, feel free to prove me wrong.)

These XMM-Newton results are nice, but tentative – they need confirming by more experiments. I can’t help but wonder how many big discoveries end up delayed or even discarded due to the cautiousness of physicists, who can scarcely believe they have found something so great. I look forward to the time when someone actually comes out and says ‘We did it – we found it.” with certainty. It would be extra nice if it were LUX. But realistically, to really convince anyone that dark matter has been found, detection via several different methods and in several different places is needed. There is a lot of work to do yet.

It’s an exciting time to be in this field, and papers like the XMM-Newton one keep us on our toes! LUX will be starting up again soon for what we hope will be a 300 day run, and an increase in sensitivity to WIMPs of around 5x. Maybe it’s time for me to re-read that paper on the axio-electric effect in liquid xenon detectors!

by Sally Shaw at October 22, 2014 04:07 PM

Tommaso Dorigo - Scientificblogging

The Quote Of The Week - Shocked And Disappointed
"Two recent results from other experiments add to the excitement of Run II. The results from Brookhaven's g-minus-two experiments with muons have a straightforward interpretation as signs of supersymmetry. The increasingly interesting results from BABAR at the Stanford Linear Accelerator Center add to the importance of B physics in Run II, and also suggest new physics. I will be shocked and disappointed if we don't have at least one major discovery."

read more

by Tommaso Dorigo at October 22, 2014 03:20 PM

Quantum Diaries

New high-speed transatlantic network to benefit science collaborations across the U.S.

This Fermilab press release came out on Oct. 20, 2014.

ESnet to build high-speed extension for faster data exchange between United States and Europe. Image: ESnet

ESnet to build high-speed extension for faster data exchange between United States and Europe. Image: ESnet

Scientists across the United States will soon have access to new, ultra-high-speed network links spanning the Atlantic Ocean thanks to a project currently under way to extend ESnet (the U.S. Department of Energy’s Energy Sciences Network) to Amsterdam, Geneva and London. Although the project is designed to benefit data-intensive science throughout the U.S. national laboratory complex, heaviest users of the new links will be particle physicists conducting research at the Large Hadron Collider (LHC), the world’s largest and most powerful particle collider. The high capacity of this new connection will provide U.S. scientists with enhanced access to data at the LHC and other European-based experiments by accelerating the exchange of data sets between institutions in the United States and computing facilities in Europe.

DOE’s Brookhaven National Laboratory and Fermi National Accelerator Laboratory—the primary computing centers for U.S. collaborators on the LHC’s ATLAS and CMS experiments, respectively—will make immediate use of the new network infrastructure once it is rigorously tested and commissioned. Because ESnet, based at DOE’s Lawrence Berkeley National Laboratory, interconnects all national laboratories and a number of university-based projects in the United States, tens of thousands of researchers from all disciplines will benefit as well.

The ESnet extension will be in place before the LHC at CERN in Switzerland—currently shut down for maintenance and upgrades—is up and running again in the spring of 2015. Because the accelerator will be colliding protons at much higher energy, the data output from the detectors will expand considerably—to approximately 40 petabytes of raw data per year compared with 20 petabytes for all of the previous lower-energy collisions produced over the three years of the LHC first run between 2010 and 2012.

The cross-Atlantic connectivity during the first successful run for the LHC experiments, which culminated in the discovery of the Higgs boson, was provided by the US LHCNet network, managed by the California Institute of Technology. In recent years, major research and education networks around the world—including ESnet, Internet2, California’s CENIC, and European networks such as DANTE, SURFnet and NORDUnet—have increased their backbone capacity by a factor of 10, using sophisticated new optical networking and digital signal processing technologies. Until recently, however, higher-speed links were not deployed for production purposes across the Atlantic Ocean—creating a network “impedance mismatch” that can harm large, intercontinental data flows.

An evolving data model
This upgrade coincides with a shift in the data model for LHC science. Previously, data moved in a more predictable and hierarchical pattern strongly influenced by geographical proximity, but network upgrades around the world have now made it possible for data to be fetched and exchanged more flexibly and dynamically. This change enables faster science outcomes and more efficient use of storage and computational power, but it requires networks around the world to perform flawlessly together.

“Having the new infrastructure in place will meet the increased need for dealing with LHC data and provide more agile access to that data in a much more dynamic fashion than LHC collaborators have had in the past,” said physicist Michael Ernst of DOE’s Brookhaven National Laboratory, a key member of the team laying out the new and more flexible framework for exchanging data between the Worldwide LHC Computing Grid centers.

Ernst directs a computing facility at Brookhaven Lab that was originally set up as a central hub for U.S. collaborators on the LHC’s ATLAS experiment. A similar facility at Fermi National Accelerator Laboratory has played this role for the LHC’s U.S. collaborators on the CMS experiment. These computing resources, dubbed Tier 1 centers, have direct links to the LHC at the European laboratory CERN (Tier 0).  The experts who run them will continue to serve scientists under the new structure. But instead of serving as hubs for data storage and distribution only among U.S.-based collaborators at Tier 2 and 3 research centers, the dedicated facilities at Brookhaven and Fermilab will be able to serve data needs of the entire ATLAS and CMS collaborations throughout the world. And likewise, U.S. Tier 2 and Tier 3 research centers will have higher-speed access to Tier 1 and Tier 2 centers in Europe.

“This new infrastructure will offer LHC researchers at laboratories and universities around the world faster access to important data,” said Fermilab’s Lothar Bauerdick, head of software and computing for the U.S. CMS group. “As the LHC experiments continue to produce exciting results, this important upgrade will let collaborators see and analyze those results better than ever before.”

Ernst added, “As centralized hubs for handling LHC data, our reliability, performance and expertise have been in demand by the whole collaboration, and now we will be better able to serve the scientists’ needs.”

An investment in science
ESnet is funded by DOE’s Office of Science to meet networking needs of DOE labs and science projects. The transatlantic extension represents a financial collaboration, with partial support coming from DOE’s Office of High Energy Physics (HEP) for the next three years. Although LHC scientists will get a dedicated portion of the new network once it is in place, all science programs that make use of ESnet will now have access to faster network links for their data transfers.

“We are eagerly awaiting the start of commissioning for the new infrastructure,” said Oliver Gutsche, Fermilab scientist and member of the CMS Offline and Computing Management Board. “After the Higgs discovery, the next big LHC milestones will come in 2015, and this network will be indispensable for the success of the LHC Run 2 physics program.”

This work was supported by the DOE Office of Science.
Fermilab is America’s premier national laboratory for particle physics and accelerator research. A U.S. Department of Energy Office of Science laboratory, Fermilab is located near Chicago, Illinois, and operated under contract by the Fermi Research Alliance, LLC. Visit Fermilab’s website at and follow us on Twitter at @FermilabToday.

Brookhaven National Laboratory is supported by the Office of Science of the U.S. Department of Energy.  The Office of Science is the single largest supporter of basic research in the physical sciences in the United States, and is working to address some of the most pressing challenges of our time.  For more information, please visit

One of ten national laboratories overseen and primarily funded by the Office of Science of the U.S. Department of Energy (DOE), Brookhaven National Laboratory conducts research in the physical, biomedical, and environmental sciences, as well as in energy technologies and national security. Brookhaven Lab also builds and operates major scientific facilities available to university, industry and government researchers. Brookhaven is operated and managed for DOE’s Office of Science by Brookhaven Science Associates, a limited-liability company founded by the Research Foundation for the State University of New York on behalf of Stony Brook University, the largest academic user of Laboratory facilities, and Battelle, a nonprofit applied science and technology organization.

Visit Brookhaven Lab’s electronic newsroom for links, news archives, graphics, and more at, follow Brookhaven Lab on Twitter,, or find us on Facebook,

The DOE Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, please visit

Media contacts:

  • Karen McNulty-Walsh, Brookhaven Media and Communications Office,, 631-344-8350
  • Kurt Riesselmann, Fermilab Office of Communication,, 630-840-3351
  • Jon Bashor, Computing Sciences Communications Manager, Lawrence Berkeley National Laboratory,, 510-486-5849

Computing contacts:

  • Lothar Bauerdick, Fermilab, US CMS software computing,, 630-840-6804
  • Oliver Gutsche, Fermilab, CMS Offline and Computing Management Board,, 630-840-8909

by Fermilab at October 22, 2014 03:15 PM

CERN Bulletin

CERN Bulletin Issue No. 43-44/2014
Link to e-Bulletin Issue No. 43-44/2014Link to all articles in this issue No.

October 22, 2014 02:58 PM

Clifford V. Johnson - Asymptotia

I Dare!
sunday_assembly_3(Click photos* for larger view) Yes. I dare to show equations during public lectures. There'll be equations in my book too. If we do not show the tools we use, how can we give a complete picture of how science works? If we keep hiding the mathematics, won't people be even more afraid of this terrifying horror we are "protecting" them from? I started my Sunday Assembly talk reflecting upon the fact that next year will make 100 years after Einstein published one of the most beautiful and far-reaching scientific works in history, General Relativity, describing how gravity works. In the first 30 seconds of the talk, I put up the equations. Just because they deserve to be seen, and to drive home the point that its not just a bunch of words, but an actual method of computation, that allows you to do quantitative science about the largest physical object we know of - the entire universe! sunday_assembly_1 It was a great audience, who seemed to enjoy the 20 minute talk as part of [...] Click to continue reading this post

by Clifford at October 22, 2014 02:29 PM

Tommaso Dorigo - Scientificblogging

ECFA Workshop: Planning For The High Luminosity LHC
I am spending a few days in Aix Les Bains, a pleasant lakeside resort in the French southwest, to follow the works of the second ECFA workshop, titled "High-Luminosity LHC". ECFA stands for "European Committee for Future Accelerators" but this particular workshop is indeed centred on the future of the LHC, despite the fact that there are at present at least half a dozen international efforts toward the design of more powerful hadron colliders, more precise linear electron-positron colliders, or still other solutions.

read more

by Tommaso Dorigo at October 22, 2014 11:04 AM

October 21, 2014

ZapperZ - Physics and Physicists

Scientific Evidence Points To A Designer?
We have had these types of anthropic universe arguments before, and I don't see this being settled anytime soon, unless we encounter an alien life form or something that dramatic.

Apparently, this physicists have been making the rounds giving talks on scientific evidence that points to a designer. Unfortunately, this claim is highly misleading. There are several issues that need to be clarified here:

1. These so-called evidence have many varying interpretations. In the hands of Stephen Hawking, he sees this as evidence that we do NOT need a designer for the universe to exist. So to claim it that they point to a designer is highly misleading, because obviously there are very smart people out there who think of the opposite.

2. Scientific evidence have varying degree of certainty. The evidence that Niobium undergoes a superconducting transition at 9.3 K is a lot more certain than many of the astrophysical parameters that we have gathered so far. It is just the nature of the study and the field.

3. It is also interesting to note that even if the claim is true, it has a significant conflict with many of the orthodox religious view of the origin of the universe, including the fact that it allows for significant time for speciation and evolution.

4. The argument that the universe has been fine-tuned for us to live in is very weak in my book. Who is there to say that if any of these parameters is different that a different type of universe couldn't appear and that different type of life forms would dominate? We are still at an infant knowledge as far as how different types of universes could form, which is one of the argument that Hawking used when he invoked the multiverse scenario. So unless that there is a convincing argument that our universe is the one and only universe that can exist, and nothing else can, then this argument falls very flat.

I find that this type of seminar can't be very productive unless there is a panel discussion presenting both sides. People who listened to this may not be aware of the holes in such arguments, and I would point out also to the any talk by those on the opposite side as well. It would have been better if they invited two scientists with opposing view, and they can show to the public how the same set of evidence leads to different conclusions. This is what happens when the full set of evidence to paint a clear picture isn't available.


by ZapperZ ( at October 21, 2014 06:38 PM

Quantum Diaries

I feel it mine

On Saturday, 4 October, Nikhef – the Dutch National Institute for Subatomic Physics where I spend long days and efforts – opened its doors, labs and facilities to the public. In addition to Nikhef, all the other institutes located in the so-called “Science Park” – the scientific district located in the east part of Amsterdam – welcomed people all day long.

It’s the second “Open Day” that I’ve attended, both as a guest and as guide. Together with my fellow theoreticians we provided answers and explanations to people’s questions and curiosities, standing in the “Big Bang Theory Corner” of the main hall. Each department in Nikhef arranged its own stand and activities, and there were plenty of things to be amazed at to cover the entire day.

The research institutes in Science Park (and outside it) offer a good overview of the concept of research, looking for what is beyond the current status of knowledge. “Verder kijken”, or looking further, is the motto of Vrije Universiteit Amsterdam, my Dutch alma mater.

I deeply like this attitude of research, the willingness to investigating what’s around the corner. As they like to define themselves, Dutch people are “future oriented”: this is manifest in several things, from the way they read the clock (“half past seven” becomes “half before eight” in Dutch) to some peculiarities of the city itself, like the presence of a lot of cultural and research institutes.

This abundance of institutes, museums, exhibitions, public libraries, music festivals, art spaces, and independent cinemas makes me feel this city as cultural place. People interact with culture in its many manifestations and are connected to it in a more dynamic way than if they were only surrounded by historical and artistic.

Back to the Open Day and Nikhef, I was pleased to see lots of people, families with kids running here and there, checking out delicate instruments with their curious hands, and groups of guys and girls (also someone who looked like he had come straight from a skate-park) stopping by and looking around as if it were their own courtyard.

The following pictures give some examples of the ongoing activities:

We had a model of the ATLAS detector built with Legos: amazing!


Copyright Nikhef

And not only toy-models. We had also true detectors, like a cloud chamber that allowed visitors to see the traces of particles passing by!


Copyright Nikhef

Weak force and anti-matter are also cool, right?


Copyright Nikhef

The majority of people here (not me) are blond and/or tall, but not tall enough to see cosmic rays with just their eyes… So, please ask the experts!


Copyright Nikhef

I think I can summarize the huge impact and the benefit of such a cool day with the words of one man who stopped by one of the experimental setups. He listened to the careful (but a bit fuzzy) explanation provided by one of the students, and said “Thanks. Now I feel it mine too.”

Many more photos are available here: enjoy!

by Andrea Signori at October 21, 2014 05:23 PM

John Baez - Azimuth

Network Theory Seminar (Part 3)


This time we use the principle of minimum power to determine what a circuit made of resistors actually does. Its ‘behavior’ is described by a functor sending circuits to linear relations between the potentials and currents at the input and output terminals. We call this the ‘black box’ functor, since it takes a circuit:

and puts a metaphorical ‘black box’ around it:

hiding the circuit’s internal details and letting us see only how it acts as viewed ‘from outside’.

For more, see the lecture notes here:

Network theory (part 32).


by John Baez at October 21, 2014 03:17 PM

Symmetrybreaking - Fermilab/SLAC

Costumes to make zombie Einstein proud

These physics-themed Halloween costume ideas are sure to entertain—and maybe even educate. Terrifying, we know.

So you haven’t picked a Halloween costume, and the big night is fast approaching. If you’re looking for something a little funny, a little nerdy and sure to impress fellow physics fans, look no further. We’ve got you covered.

1. Dark energy

This is an active costume, perfect for the party-goer who plans to consume a large quantity of sugar. Suit up in all black or camouflage, then spend your evening squeezing between people and pushing them apart.

Congratulations! You’re dark energy: a mysterious force causing the accelerating expansion of the universe, intriguing in the lab and perplexing on the dance floor.

2. Cosmic inflation

Theory says that a fraction of a second after the big bang, the universe grew exponentially, expanding so that tiny fluctuations were stretched into the seeds of entire galaxies.

But good luck getting that costume through the door.

Instead, take a simple yellow life vest and draw the cosmos on it: stars, planets, asteroids, whatever you fancy. When friends pull on the emergency tab, the universe will grow.

3. Heisenberg Uncertainty Principle

Here’s a great excuse to repurpose your topical Breaking Bad costume from last year.

Walter White—aka “Heisenberg”—may have been a chemistry teacher, but the Heisenberg Uncertainty Principle is straight out of physics. Named after Werner Heisenberg, a German physicist credited with the creation of quantum mechanics, the Heisenberg Uncertainty Principle states that the more accurately you know the position of a particle, the less information you know about its momentum.

Put on Walter White’s signature hat and shades (or his yellow suit and respirator), but then add some uncertainty by pasting Riddler-esque question marks to your outfit.

4. Bad neutrino

A warning upfront: Only the ambitious and downright extroverted should attempt this costume.

Neutrinos are ghostly particles that pass through most matter undetected. In fact, trillions of neutrinos pass through your body every second without your knowledge.

But you aren’t going to go as any old neutrino. Oh no. You’re a bad neutrino—possibly the worst one in the universe—so you run into everything: lampposts, trees, haunted houses and yes, people. Don a simple white sheet and spend the evening interacting with everyone and everything.

5. Your favorite physics experiment

You physics junkies know that there are a lot of experiments with odd acronyms and names that are ripe for Halloween costumes. You can go as ATLAS (experiment at the Large Hadron Collider / character from Greek mythology), DarkSide (dark matter experiment at Gran Sasso National Laboratory / good reason to repurpose your Darth Vader costume), PICASSO (dark matter experiment at SNOLAB / creator of Cubism), MINERvA (Fermilab neutrino experiment / Roman goddess of wisdom), or the Dark Energy Survey (dark energy camera located at the Blanco Telescope in Chile / good opportunity for a pun).

Physics-loving parents can go as explorer Daniel Boone, while the kids go as neutrino experiments MicroBooNE and MiniBooNE. The kids can wear mini fur hats of their own or dress as detector tanks to be filled with candy.

6. Feynman diagram

You might know that a Feynman diagram is a drawing that uses lines and squiggles to represent a particle interaction. But have you ever noticed that they sometimes look like people? Try out this new take on the black outfit/white paint skeleton costume. Bonus points for going as a penguin diagram.

7. Antimatter

Break out the bell-bottoms and poster board. In bold letters, scrawl the words of your choosing: “I hate things!,” “Stuff is awful!,” and “Down with quarks!” will all do nicely. Protest from house to house and declare with pride that you are antimatter. It’s a fair critique: Physicists still aren’t sure why matter dominates the universe when equal amounts of matter and antimatter should have been created in the big bang.

Fortunately, you don’t have to solve this particular puzzle on your quest for candy. Just don’t high five anyone; you might annihilate.

8. Entangled particles

Einstein described quantum entanglement as “spooky action at a distance”—the perfect costume for Halloween. Entangled particles are extremely strange. Measuring one automatically determines the state of the other, instantaneously.

Find someone you are extremely in tune with and dress in opposite colors, like black and white. When no one is observing you, you can relax. But when interacting with people, be sure to coordinate movements. They spin to the left, you spin to the right. They wave with the right hand? You wave with the left. You get the drill.

You can also just wrap yourselves together in a net. No one said quantum entanglement has to be hard.

9. Holographic you(niverse)

The universe may be like a hologram, according to a theory currently being tested at Fermilab’s Holometer experiment. If so, information about spacetime is chunked into 2-D bits that only appear three-dimensional from our perspective.

Help others imagine this bizarre concept by printing out a photo of yourself and taping it to your front. You’ll still technically be 3-D, but that two-dimensional picture of your face will still start some interesting discussions. Perhaps best not to wear this if you have a busy schedule or no desire to discuss the nature of time and space while eating a Snickers.

10. Your favorite particle

There are many ways to dress up as a fundamental particle. Bring a lamp along to trick-or-treat to go as the photon, carrier of light. Hand out cookies to go as the Higgs boson, giver of mass. Spend the evening attaching things to people to go as a gluon.

To branch out beyond the Standard Model of particle physics, go as a supersymmetric particle, or sparticle: Wear a gladiator costume and shout, “I am Sparticle!” whenever someone asks about your costume.

Or grab a partner to become a meson, a particle made of a quark and antiquark. Mesons are typically unstable, so whenever you unlink arms, be sure to decay in a shower of electrons and neutrinos—or candy corn.


Like what you see? Sign up for a free subscription to symmetry!

by Lauren Biron at October 21, 2014 02:51 PM

Jester - Resonaances

Dark matter or pulsars? AMS hints it's neither.
Yesterday AMS-02 updated their measurement of cosmic-ray positron and electron fluxes. The newly published data extend to positron energies 500 GeV, compared to 350 GeV in the previous release. The central value of the positron fraction in the highest energy bin is one third of the error bar lower than the central value of the next-to-highestbin.  This allows the collaboration to conclude that the positron fraction has a maximum and starts to decrease at high energies :]  The sloppy presentation and unnecessary hype obscures the fact that AMS actually found something non-trivial.  Namely, it is interesting that the positron fraction, after a sharp rise between 10 and 200 GeV, seems to plateau at higher energies at the value around 15%.  This sort of behavior, although not expected by popular models of cosmic ray propagation, was actually predicted a few years ago, well before AMS was launched.  

Before I get to the point, let's have a brief summary. In 2008 the PAMELA experiment observed a steep rise of the cosmic ray positron fraction between 10 and 100 GeV. Positrons are routinely produced by scattering of high energy cosmic rays (secondary production), but the rise was not predicted by models of cosmic ray propagations. This prompted speculations of another (primary) source of positrons: from pulsars, supernovae or other astrophysical objects, to  dark matter annihilation. The dark matter explanation is unlikely for many reasons. On the theoretical side, the large annihilation cross section required is difficult to achieve, and it is difficult to produce a large flux of positrons without producing an excess of antiprotons at the same time. In particular, the MSSM neutralino entertained in the last AMS paper certainly cannot fit the cosmic-ray data for these reasons. When theoretical obstacles are overcome by skillful model building, constraints from gamma ray and radio observations disfavor the relevant parameter space. Even if these constraints are dismissed due to large astrophysical uncertainties, the models poorly fit the shape the electron and positron spectrum observed by PAMELA, AMS, and FERMI (see the addendum of this paper for a recent discussion). Pulsars, on the other hand, are a plausible but handwaving explanation: we know they are all around and we know they produce electron-positron pairs in the magnetosphere, but we cannot calculate the spectrum from first principles.

But maybe primary positron sources are not needed at all? The old paper by Katz et al. proposes a different approach. Rather than starting with a particular propagation model, it assumes the high-energy positrons observed by PAMELA are secondary, and attempts to deduce from the data the parameters controlling the propagation of cosmic rays. The logic is based on two premises. Firstly, while production of cosmic rays in our galaxy contains many unknowns, the production of different particles is strongly correlated, with the relative ratios depending on nuclear cross sections that are measurable in laboratories. Secondly, different particles propagate in the magnetic field of the galaxy in the same way, depending only on their rigidity (momentum divided by charge). Thus, from an observed flux of one particle, one can predict the production rate of other particles. This approach is quite successful in predicting the cosmic antiproton flux based on the observed boron flux. For positrons, the story is more complicated because of large energy losses (cooling) due to synchrotron and inverse-Compton processes. However, in this case one can make the  exercise of computing the positron flux assuming no losses at all. The result correspond to roughly 20% positron fraction above 100 GeV. Since in the real world cooling can only suppress the positron flux, the value computed assuming no cooling represents an upper bound on the positron fraction.

Now, at lower energies, the observed positron flux is a factor of a few below the upper bound. This is already intriguing, as hypothetical primary positrons could in principle have an arbitrary flux,  orders of magnitude larger or smaller than this upper bound. The rise observed by PAMELA can be interpreted that the suppression due to cooling decreases as positron energy increases. This is not implausible: the suppression depends on the interplay of the cooling time and mean propagation time of positrons, both of which are unknown functions of energy. Once the cooling time exceeds the propagation time the suppression factor is completely gone. In such a case the positron fraction should saturate the upper limit. This is what seems to be happening at the energies 200-500 GeV probed by AMS, as can be seen in the plot. Already the previous AMS data were consistent with this picture, and the latest update only strengthens it.

So, it may be that the mystery of cosmic ray positrons has a simple down-to-galactic-disc explanation. If further observations show the positron flux climbing  above the upper limit or dropping suddenly, then the secondary production hypothesis would be invalidated. But, for the moment, the AMS data seems to be consistent with no primary sources, just assuming that the cooling time of positrons is shorter than predicted by the state-of-the-art propagation models. So, instead of dark matter, AMS might have discovered models of cosmic-ray propagation need a fix. That's less spectacular, but still worthwhile.

Thanks to Kfir for the plot and explanations. 

by Jester ( at October 21, 2014 08:49 AM

October 20, 2014

John Baez - Azimuth

Network Theory (Part 32)

Okay, today we will look at the ‘black box functor’ for circuits made of resistors. Very roughly, this takes a circuit made of resistors with some inputs and outputs:

and puts a ‘black box’ around it:

forgetting the internal details of the circuit and remembering only how the it behaves as viewed from outside. As viewed from outside, all the circuit does is define a relation between the potentials and currents at the inputs and outputs. We call this relation the circuit’s behavior. Lots of different choices of the resistances R_1, \dots, R_6 would give the same behavior. In fact, we could even replace the whole fancy circuit by a single edge with a single resistor on it, and get a circuit with the same behavior!

The idea is that when we use a circuit to do something, all we care about is its behavior: what it does as viewed from outside, not what it’s made of.

Furthermore, we’d like the behavior of a system made of parts to depend in a simple way on the external behaviors of its parts. We don’t want to have to ‘peek inside’ the parts to figure out what the whole will do! Of course, in some situations we do need to peek inside the parts to see what the whole will do. But in this particular case we don’t—at least in the idealization we are considering. And this fact is described mathematically by saying that black boxing is a functor.

So, how do circuits made of resistors behave? To answer this we first need to remember what they are!


Remember that for us, a circuit made of resistors is a mathematical structure like this:

It’s a cospan where:

\Gamma is a graph labelled by resistances. So, it consists of a finite set N of nodes, a finite set E of edges, two functions

s, t : E \to N

sending each edge to its source and target nodes, and a function

r : E \to (0,\infty)

that labels each edge with its resistance.

i: I \to \Gamma is a map of graphs labelled by resistances, where I has no edges. A labelled graph with no edges has nothing but nodes! So, the map i is just a trick for specifying a finite set of nodes called inputs and mapping them to N. Thus i picks out some nodes of \Gamma and declares them to be inputs. (However, i may not be one-to-one! We’ll take advantage of that subtlety later.)

o: O \to \Gamma is another map of graphs labelled by resistances, where O again has no edges, and we call its nodes outputs.

The principle of minimum power

So what does a circuit made of resistors do? This is described by the principle of minimum power.

Recall from Part 27 that when we put it to work, our circuit has a current I_e flowing along each edge e \in E. This is described by a function

I: E \to \mathbb{R}

It also has a voltage across each edge. The word ‘across’ is standard here, but don’t worry about it too much; what matters is that we have another function

V: E \to \mathbb{R}

describing the voltage V_e across each edge e.

Resistors heat up when current flows through them, so they eat up electrical power and turn this power into heat. How much? The power is given by

\displaystyle{ P = \sum_{e \in E} I_e V_e }

So far, so good. But what does it mean to minimize power?

To understand this, we need to manipulate the formula for power using the laws of electrical circuits described in Part 27. First, Ohm’s law says that for linear resistors, the current is proportional to the voltage. More precisely, for each edge e \in E,

\displaystyle{ I_e = \frac{V_e}{r_e} }

where r_e is the resistance of that edge. So, the bigger the resistance, the less current flows: that makes sense. Using Ohm’s law we get

\displaystyle{ P = \sum_{e \in E} \frac{V_e^2}{r_e} }

Now we see that power is always nonnegative! Now it makes more sense to minimize it. Of course we could minimize it simply by setting all the voltages equal to zero. That would work, but that would be boring: it gives a circuit with no current flowing through it. The fun starts when we minimize power subject to some constraints.

For this we need to remember another law of electrical circuits: a spinoff of Kirchhoff’s voltage law. This says that we can find a function called the potential

\phi: N \to \mathbb{R}

such that

V_e = \phi_{s(e)} - \phi_{t(e)}

for each e \in E. In other words, the voltage across each edge is the difference of potentials at the two ends of this edge.

Using this, we can rewrite the power as

\displaystyle{ P = \sum_{e \in E} \frac{1}{r_e} (\phi_{s(e)} - \phi_{t(e)})^2 }

Now we’re really ready to minimize power! Our circuit made of resistors has certain nodes called terminals:

T \subseteq N

These are the nodes that are either inputs or outputs. More precisely, they’re the nodes in the image of

i: I \to \Gamma


o: O \to \Gamma

The principle of minimum power says that:

If we fix the potential \phi on all terminals, the potential at other nodes will minimize the power

\displaystyle{ P(\phi) = \sum_{e \in E} \frac{1}{r_e} (\phi_{s(e)} - \phi_{t(e)})^2 }

subject to this constraint.

This should remind you of all the other minimum or maximum principles you know, like the principle of least action, or the way a system in thermodynamic equilibrium maximizes its entropy. All these principles—or at least, most of them—are connected. I could talk about this endlessly. But not now!

Now let’s just use the principle of minimum power. Let’s see what it tells us about the behavior of an electrical circuit.

Let’s imagine changing the potential \phi by adding some multiple of a function

\psi: N \to \mathbb{R}

If this other function vanishes at the terminals:

\forall n \in T \; \; \psi(n) = 0

then \phi + x \psi doesn’t change at the terminals as we change the number x.

Now suppose \phi obeys the principle of minimum power. In other words, supposes it minimizes power subject to the constraint of taking the values it does at the terminals. Then we must have

\displaystyle{ \frac{d}{d x} P(\phi + x \psi)\Big|_{x = 0} }


\forall n \in T \; \; \psi(n) = 0

This is just the first derivative test for a minimum. But the converse is true, too! The reason is that our power function is a sum of nonnegative quadratic terms. Its graph will look like a paraboloid. So, the power has no points where its derivative vanishes except minima, even when we constrain \phi by making it lie on a linear subspace.

We can go ahead and start working out the derivative:

\displaystyle{ \frac{d}{d x} P(\phi + x \psi)! = ! \frac{d}{d x} \sum_{e \in E} \frac{1}{r_e} (\phi_{s(e)} - \phi_{t(e)} + x(\psi_{s(e)} -\psi_{t(e)}))^2  }

To work out the derivative of these quadratic terms at x = 0, we only need to keep the part that’s proportional to x. The rest gives zero. So:

\begin{array}{ccl} \displaystyle{ \frac{d}{d t} P(\phi + x \psi)\Big|_{x = 0} } &=& \displaystyle{ \frac{d}{d x} \sum_{e \in E} \frac{x}{r_e} (\phi_{s(e)} - \phi_{t(e)}) (\psi_{s(e)} - \psi_{t(e)}) \Big|_{x = 0} } \\ \\  &=&   \displaystyle{  \sum_{e \in E} \frac{1}{r_e} (\phi_{s(e)} - \phi_{t(e)}) (\psi_{s(e)} - \psi_{t(e)}) }  \end{array}

The principle of minimum power says this is zero whenever \psi : N \to \mathbb{R} is a function that vanishes at terminals. By linearity, it’s enough to consider functions \psi that are zero at every node except one node n that is not a terminal. By linearity we can also assume \psi(n) = 1.

Given this, the only nonzero terms in the sum

\displaystyle{ \sum_{e \in E} \frac{1}{r_e} (\phi_{s(e)} - \phi_{t(e)}) (\psi_{s(e)} - \psi_{t(e)}) }

will be those involving edges whose source or target is n. We get

\begin{array}{ccc} \displaystyle{ \frac{d}{d x} P(\phi + x \psi)\Big|_{x = 0} } &=& \displaystyle{ \sum_{e: \; s(e) = n}  \frac{1}{r_e} (\phi_{s(e)} - \phi_{t(e)})}  \\  \\        && -\displaystyle{ \sum_{e: \; t(e) = n}  \frac{1}{r_e} (\phi_{s(e)} - \phi_{t(e)}) }   \end{array}

So, the principle of minimum power says precisely

\displaystyle{ \sum_{e: \; s(e) = n}  \frac{1}{r_e} (\phi_{s(e)} - \phi_{t(e)}) = \sum_{e: \; t(e) = n}  \frac{1}{r_e} (\phi_{s(e)} - \phi_{t(e)}) }

for all nodes n that aren’t terminals.

What does this mean? You could just say it’s a set of linear equations that must be obeyed by the potential \phi. So, the principle of minimum power says that fixing the potential at terminals, the potential at other nodes must be chosen in a way that obeys a set of linear equations.

But what do these equations mean? They have a nice meaning. Remember, Kirchhoff’s voltage law says

V_e = \phi_{s(e)} - \phi_{t(e)}

and Ohm’s law says

\displaystyle{ I_e = \frac{V_e}{r_e} }

Putting these together,

\displaystyle{ I_e = \frac{1}{r_e} (\phi_{s(e)} - \phi_{t(e)}) }

so the principle of minimum power merely says that

\displaystyle{ \sum_{e: \; s(e) = n} I_e = \sum_{e: \; t(e) = n}  I_e }

for any node n that is not a terminal.

This is Kirchhoff’s current law: for any node except a terminal, the total current flowing into that node must equal the total current flowing out! That makes a lot of sense. We allow current to flow in or out of our circuit at terminals, but ‘inside’ the circuit charge is conserved, so if current flows into some other node, an equal amount has to flow out.

In short: the principle of minimum power implies Kirchoff’s current law! Conversely, we can run the whole argument backward and derive the principle of minimum power from Kirchhoff’s current law. (In both the forwards and backwards versions of this argument, we use Kirchhoff’s voltage law and Ohm’s law.)

When the node n is a terminal, the quantity

\displaystyle{  \sum_{e: \; s(e) = n} I_e \; - \; \sum_{e: \; t(e) = n}  I_e }

need not be zero. But it has an important meaning: it’s the amount of current flowing into that terminal!

We’ll call this I_n, the current at the terminal n \in T. This is something we can measure even when our circuit has a black box around it:

So is the potential \phi_n at the terminal n. It’s these currents and potentials at terminals that matter when we try to describe the behavior of a circuit while ignoring its inner workings.

Black boxing

Now let me quickly sketch how black boxing becomes a functor.

A circuit made of resistors gives a linear relation between the potentials and currents at terminals. A relation is something that can hold or fail to hold. A ‘linear’ relation is one defined using linear equations.

A bit more precisely, suppose we choose potentials and currents at the terminals:

\psi : T \to \mathbb{R}

J : T \to \mathbb{R}

Then we seek potentials and currents at all the nodes and edges of our circuit:

\phi: N \to \mathbb{R}

I : E \to \mathbb{R}

that are compatible with our choice of \psi and J. Here compatible means that

\psi_n = \phi_n


J_n = \displaystyle{  \sum_{e: \; s(e) = n} I_e \; - \; \sum_{e: \; t(e) = n}  I_e }

whenever n \in T, but also

\displaystyle{ I_e = \frac{1}{r_e} (\phi_{s(e)} - \phi_{t(e)}) }

for every e \in E, and

\displaystyle{  \sum_{e: \; s(e) = n} I_e \; = \; \sum_{e: \; t(e) = n}  I_e }

whenever n \in N - T. (The last two equations combine Kirchoff’s laws and Ohm’s law.)

There either exist I and \phi making all these equations true, in which case we say our potentials and currents at the terminals obey the relation… or they don’t exist, in which case we say the potentials and currents at the terminals don’t obey the relation.

The relation is clearly linear, since it’s defined by a bunch of linear equations. With a little work, we can make it into a linear relation between potentials and currents in

\mathbb{R}^I \oplus \mathbb{R}^I

and potentials and currents in

\mathbb{R}^O \oplus \mathbb{R}^O

Remember, I is our set of inputs and O is our set of outputs.

In fact, this process of getting a linear relation from a circuit made of resistors defines a functor:

\blacksquare : \mathrm{ResCirc} \to \mathrm{LinRel}

Here \mathrm{ResCirc} is the category where morphisms are circuits made of resistors, while \mathrm{LinRel} is the category where morphisms are linear relations.

More precisely, here is the category \mathrm{ResCirc}:

• an object of \mathrm{ResCirc} is a finite set;

• a morphism from I to O is an isomorphism class of circuits made of resistors:

having I as its set of inputs and O as its set of outputs;

• we compose morphisms in \mathrm{ResCirc} by composing isomorphism classes of cospans.

(Remember, circuits made of resistors are cospans. This lets us talk about isomorphisms between them. If you forget the how isomorphism between cospans work, you can review it in Part 31.)

And here is the category \mathrm{LinRel}:

• an object of \mathrm{LinRel} is a finite-dimensional real vector space;

• a morphism from U to V is a linear relation R \subseteq U \times V, meaning a linear subspace of the vector space U \times V;

• we compose a linear relation R \subseteq U \times V and a linear relation S \subseteq V \times W in the usual way we compose relations, getting:

SR = \{(u,w) \in U \times W : \; \exists v \in V \; (u,v) \in R \mathrm{\; and \;} (v,w) \in S \}

Next steps

So far I’ve set up most of the necessary background but not precisely defined the black boxing functor

\blacksquare : \mathrm{ResCirc} \to \mathrm{LinRel}

There are some nuances I’ve glossed over, like the difference between inputs and outputs as elements of I and O and their images in N. If you want to see the precise definition and the proof that it’s a functor, read our paper:

• John Baez and Brendan Fong, A compositional framework for passive linear networks.

The proof is fairly long: there may be a much quicker one, but at least this one has the virtue of introducing a lot of nice ideas that will be useful elsewhere.

Perhaps next time I will clarify the nuances by doing an example.

by John Baez at October 20, 2014 10:00 PM

ATLAS Experiment

Defending Your Life (Part 2)

I’ve been working on our simulation software for a long time, and I’m often asked “what on earth is that?” This is my attempt to help you love simulation as much as I do. This is a follow up to Part 1, which told you all about the first step of good simulation software, called “event generation”. In that step, we had software that gave us a list of stable particles that our detector might be able to see. And we’re trying to find some “meons” that our friend the theorist dreamed up.

One little problem with those wonderful event generators is that they don’t know anything about our experiment, ATLAS. We need a different piece of software to take those particles and move them through the detector one by one, helping model the detector’s response to each one of the particles as it goes. There are a few pieces of software that can do that, but the one that we use most is called Geant4. Geant4 is publicly available, and is described as a “toolkit” on their webpage. What that means is that it knows about basic concepts, but it doesn’t do specifics. Like building a giant lego house out of a bag of bricks, you have to figure out what fits where, and often throw out things that don’t fit.

One of the detector layouts that we simulate

The first part of a good detector simulation is the detector description. Every piece of the detector has to be put together, with the right material assigned to each. We have a detector description with over five million (!) volumes and about 400 different materials (from Xenon to Argon to Air to Aerogel and Kapton Cable). There are a few heroes of ATLAS who spend a lot of time taking technical drawings (and photographs, because the technical drawings aren’t always right!) of the detector and translating them into something Geant4 can use. You can’t put every wire and pipe in – the simulation would take an eternity! – so you have to find shortcuts sometimes. It’s a painstaking process that’s still ongoing today. We continuously refine and improve our description, adding pieces that weren’t important at the beginning several years ago but are starting to be important now (like polyboron neutron shielding in our forward region; few people thought early on that we would be able to model low-energy neutron flux in our detector with Geant4, because it’s really complex nuclear physics, but we’re getting so close to being able to do so that we’ve gone back to re-check that our materials’ neutron capture properties are correct). And sometimes we go back and revise things that were done approximately in the beginning because we think we can do better. This part also involves making a detailed magnetic field map. We can’t measure the field everywhere in the detector (like deep in the middle of the calorimeter), and it takes too much time to constantly simulate the currents flowing through the magnets and their effect on the particles moving through the detector, so we do that simulation once and save the magnetic field that results.

A simulated black hole event. But what do meons look like?

Next is a good set of physics models. Geant4 has a whole lot of them that you can use and (fortunately!) they have a default that works pretty well for us. Those physics models describe each process (the photoelectric effect, Compton scattering, bremsstrahlung, ionization, multiple scattering, decays, nuclear interactions, etc) for each particle. Some are very, very complicated, as you can probably imagine. You have to choose, at this point, what physics you’re interested in. Geant4 can be used for simulation of space, simulation of cells and DNA, and simulations of radioactive environments. If we used the most precise models for everything, our simulation would never finish running! Instead, we take the fastest model whose results we can’t really distinguish from the most detailed models. That is, we turn off everything that we don’t really notice in our detector anyway. Sometimes we don’t get that right and have to go back and adjust things further – but usually we’ve erred on the side of a slower, more accurate simulation.

The last part is to “teach” Geant4 what you want to save. All Geant4 cares about is particles and materials – it doesn’t inherently know the difference between some silicon that is a part of a computer chip somewhere in the detector and the silicon that makes up the sensors in much of our inner detector. So we have to say “these are the parts of the detector that we care about most” (called “sensitive” detectors). There are a lot of technical tricks to optimizing the storage, but in the end we want to write files with all the little energy deposits that Geant4 has made, their time and location – and sometimes information (that we call “truth”) about what really happened in the simulation, so later we can find out how good our reconstruction software was at correctly identifying photons and their conversions into electron-positron pairs, for example.

The fun part of working on the simulation software is that you have to learn everything about the experiment. You have to know how much time after the interaction every piece of the detector is sensitive, so that you can avoid wasting time simulating particles long after that time. You get to learn when things were installed incorrectly or are misaligned, because you need those effects in the simulation. When people want to upgrade a part of the detector, you have to learn what they have in mind, and then (often) help them think of things they haven’t dealt with yet that might affect other parts of the detector (like cabling behind their detector, which we often have to think hard about). You also have to know about the physics that each detector is sensitive to, what approximations are reasonable, and what approximations you’re already making that they might need to check on.

That also brings us back to our friend’s meons. If they decay very quickly into Standard Model particles, then the event generator will do all the hard work. But if they stick around long enough to interact with the detector, then we have to ask our friend for a lot more information, like how they interact with different materials. For some funny theoretical particles like magnetic monopoles, R-hadrons, and stable charginos, we have to write our own Geant4 physics modules, with a lot of help from theorists.

The detector simulation is a great piece of software to work on – but that’s not the end of it! After the simulation comes the final step, “digitization”, which I’ll talk about next time – and we’ll find out the fate of our buddy’s meon theory.

ZachMarshall Zach Marshall is a Divisional Fellow at the Lawrence Berkeley National Laboratory in California. His research is focused on searches for supersymmetry and jet physics, with a significant amount of time spent working on software and trying to help students with physics and life in ATLAS.

by Zachary Marshall at October 20, 2014 03:21 PM

Jester - Resonaances

Weekend Plot: Bs mixing phase update
Today's featured plot was released last week by the LHCb collaboration:

It shows the CP violating phase in Bs meson mixing, denoted as φs,  versus the difference of the decay widths between the two Bs meson eigenstates. The interest in φs comes from the fact that it's  one of the precious observables that 1) is allowed by the symmetries of the Standard Model, 2) is severely suppressed due to the CKM structure of flavor violation in the Standard Model. Such observables are a great place to look for new physics (other observables in this family include Bs/Bd→μμ, K→πνν, ...). New particles, even too heavy to be produced directly at the LHC, could produce measurable contributions to φs as long as they don't respect the Standard Model flavor structure. For example, a new force carrier with a mass as large as 100-1000 TeV and order 1 flavor- and CP-violating coupling to b and s quarks would be visible given the current experimental precision. Similarly, loops of supersymmetric particles with 10 TeV masses could show up, again if the flavor structure in the superpartner sector is not aligned with that in the  Standard Model.

The phase φs can be measured in certain decays of neutral Bs mesons where the process involves an interference of direct decays and decays through oscillation into the anti-Bs meson. Several years ago measurements at Tevatron's D0 and CDF experiments suggested a large new physics contribution. The mild excess has gone away since, like many other such hints.  The latest value quoted by LHCb is φs = - 0.010 ± 0.040, which combines earlier measurements of the Bs → J/ψ π+ π- and  Bs → Ds+ Ds- decays with  the brand new measurement of the Bs → J/ψ K+ K- decay. The experimental precision is already comparable to the Standard Model prediction of φs = - 0.036. Further progress is still possible, as the Standard Model prediction can be computed to a few percent accuracy.  But the room for new physics here is getting tighter and tighter.

by Jester ( at October 20, 2014 02:20 PM

Symmetrybreaking - Fermilab/SLAC

Transatlantic data-transfer gets a boost

New links will improve the flow of data from the Large Hadron Collider to US institutions.

Scientists across the US will soon have access to new, ultra high-speed network links spanning the Atlantic Ocean.

A new project is currently underway to extend the US Department of Energy’s Energy Sciences Network, or ESnet, to London, Amsterdam and Geneva.

Although the project is designed to benefit data-intensive science throughout the US national laboratory complex, heaviest users of the new links will be particle physicists conducting research at the Large Hadron Collider, the world’s largest and most powerful particle collider. The high capacity of this new connection will provide US-based scientists with enhanced access to data at the LHC and other European-based experiments by accelerating the exchange of data sets between institutions in the US and computing facilities in Europe.

“After the Higgs discovery, the next big LHC milestones will come in 2015,” says Oliver Gutsche, Fermilab scientist and member of the CMS Offline and Computing Management Board. “And this network will be indispensable for the success of the [next LHC physics program].”

DOE’s Brookhaven National Laboratory and Fermi National Accelerator Laboratory—the primary computing centers for US collaborators on the LHC’s ATLAS and CMS experiments, respectively—will make immediate use of the new network infrastructure, once it is rigorously tested and commissioned. Because ESnet, based at DOE’s Lawrence Berkeley National Laboratory, interconnects all national laboratories and a number of university-based projects in the US, tens of thousands of researchers from other disciplines will benefit as well. 

The ESnet extension will be in place before the LHC at CERN in Switzerland—currently shut down for maintenance and upgrades—is up and running again in the spring of 2015. Because the accelerator will be colliding protons at much higher energy, the data output from the detectors will expand considerably to approximately 40 petabytes of RAW data per year, compared with 20 petabytes for all of the previous lower-energy collisions produced over the three years of the LHC’s first run between 2010 and 2012.

The cross-Atlantic connectivity during the first successful run for the LHC experiments was provided by the US LHCNet network, managed by the California Institute of Technology. In recent years, major research and education networks around the world—including ESnet, Internet2, California’s CENIC, and European networks such as DANTE, SURFnet and NORDUnet—have increased their backbone capacity by a factor of 10, using sophisticated new optical networking and digital signal processing technologies. Until recently, however, higher-speed links were not deployed for production purposes across the Atlantic Ocean. 

Courtesy of: Brookhaven/Fermilab

An evolving data model

This upgrade coincides with a shift in the data model for LHC science. Previously, data moved in a more predictable and hierarchical pattern strongly influenced by geographical proximity, but network upgrades around the world have now made it possible for data to be fetched and exchanged more flexibly and dynamically. This change enables faster science outcomes and more efficient use of storage and computational power, but it requires networks around the world to perform flawlessly together. 

“Having the new infrastructure in place will meet the increased need for dealing with LHC data and provide more agile access to that data in a much more dynamic fashion than LHC collaborators have had in the past,” says physicist Michael Ernst of Brookhaven National Laboratory, a key member of the team laying out the new and more flexible framework for exchanging data between the Worldwide LHC Computing Grid centers. 

Ernst directs a computing facility at Brookhaven Lab that was originally set up as a central hub for US collaborators on the LHC’s ATLAS experiment. A similar facility at Fermi National Accelerator Laboratory has played this role for the LHC’s US collaborators on the CMS experiment. These computing resources, dubbed “Tier 1” centers, have direct links to the LHC at Europe’s CERN laboratory (Tier 0).

The experts who run them will continue to serve scientists under the new structure. But instead of serving only as hubs for data storage and distribution among US-based collaborators at Tier 2 and 3 research centers, the dedicated facilities at Brookhaven and Fermilab will also be able to serve data needs of the entire ATLAS and CMS collaborations throughout the world. And likewise, US Tier 2 and Tier 3 research centers will have higher-speed access to Tier 1 and Tier 2 centers in Europe. 

“This new infrastructure will offer LHC researchers at laboratories and universities around the world faster access to important data," says Fermilab’s Lothar Bauerdick, head of software and computing for the US CMS group. "As the LHC experiments continue to produce exciting results, this important upgrade will let collaborators see and analyze those results better than ever before.”

Ernst adds, “As centralized hubs for handling LHC data, our reliability, performance, and expertise have been in demand by the whole collaboration and now we will be better able to serve the scientists’ needs.”

Fermilab published a version of this article as a press release.


Like what you see? Sign up for a free subscription to symmetry!

October 20, 2014 01:00 PM