146 points by kzrdude on Aug 1, 2022 | hide | past | favorite | 49 comments

 I finished up my PhD in scattering amplitudes recently. I checked out the abstract of this paper, and it certainly looks like an interesting result and will likely be the basis for new results in the field. For perspective, I estimate that there are a few such results in the field per year which could have roughly the same quanta magazine article written about them.
 Hey, I know that this is probably too broad a question, but I'm curious: what kind of mathematics did you require to do your doctorate? I'm curious as to what I'd need to (re)-learn to truly understand this.
 You didn't ask me, but I can shed some light, having earned a physics PhD in the early 90s, after doing an undergrad double major in math and physics.The undergrad physics curriculum includes the standard college sequence of calculus and differential equations. Many physics students take more math than this, or today, probably some computer science. We took a "physics" course called "theoretical physics" that filled some gaps in the math sequence, most notably multivariate calculus in cylindrical and spherical coordinates, path integrals, and some stuff like that. Specific upper level physics courses also introduced math topics as needed.Graduate coursework re-iterated the same "theoretical physics" course, but it was called "mathematical methods." We used a textbook by Arfken, that covered the same multivariate calculus plus vector and tensor math. Finding out what the current textbook du jour is for this course, and perusing the table of contents, would give you a pretty good idea of what to expect.I didn't go into scattering theory as a specialty, but we covered it in some of the other courses, and I remember doing difficult integrals and differential equations in spherical polar coordinates. The coursework quickly runs you out of problems that can be solved in closed form. I'd be shocked if the real work isn't done today with mostly numerical techniques.
 MengerSponge on Aug 2, 2022 https://webspace.science.uu.nl/~gadda001/goodtheorist/index....You need to work up to (and through) Quantum Field Theory to understand this.Thankfully, you don't need to be as good or as thorough as Gerard 't Hooft as you learn all those things!
 joe__f on Aug 2, 2022 Scattering Amplitudes in Gauge Theory and Gravity by Elvang and Huang is a standard introduction to the field, if you wanted to take a look and see what you might need to learn to get up to scratch. There are a lot of prerequisites.
 Any possibility of these prereqs being listed?
 You need a good foundation in quantum field theory to be able to read Elvang and Huang effectively. 'Peskin and Schroeder' and Srednicki are good standard introductions to QFT. If you understood around half of either of those text books then I think you would be in a good position.QFT itself has many prerequisites; you need to be well versed in special relativity, electromagnetism, classical field theory and quantum mechanics to get a hold on it. Each of these areas have their own mathematical prerequisites, which include algebra and calculus, multivariable calculus, pdes, complex analysis and fourier analysis.General relativity and string theory are also comparative to QFT in scope, prerequisite knowledge and utility in other areas of mathematical physics. You would need to learn these if you wanted to start reading a bit more broadly in scattering amplitudes or in another part of math phys, but you can go quite a long way in scattering amplitudes without too much of them.
 Thank you. My formal physics education ends with QM, so that seems very much within my reach of intellectual curiosity.As someone who has taught myself significantly more than formal education could, I know how valuable it is to have this sort of knowledge trail accessible. Much appreciated.
 Any guesses what this could stem from?
 I think it's a bit of a feature of quanta magazine's approach to science journalism (and probably the current zeitgeist in science journalism in general). If you want to make work like this accessible then it's necessary for a large portion of the (pop science) article to be covering topics that are common to most of the recent work in the field, with a lot of the same analogies and motivations. I think it's nice that quanta magazine are trying to make this kind of work accessible and interesting for people, but I feel that their approach sacrifices perspective on how relevant that work is, which is why I tried to add some here.
 I agree with everything here, but suspect the person you replied to was asking what caused the strange correspondence :P (or, even if they weren't, I'd like to ask for myself)
 mmazing on Aug 2, 2022 Thanks for the response, but I was mostly asking about your experience as someone working in this particular field.Do you have any guesses for what might be the connection between these two different scattering amplitudes in seemingly disconnected physical phenomenon?
 I read the paper more carefully and had a think about this, and here are my thoughts;Dixon et al. find a duality linking form factors and scattering amplitudes. Scattering amplitudes are written as a 'loop expansion', which you could think of as a bit like a Taylor expansion, and I think a remarkable feature of their result is that it appears to hold for all orders of this expansion. I've seen some results relating form factors and scattering amplitudes at only the first order of this expansion before.Their result is based on a mathematical observation that the functions that the scattering amplitudes are built from at each loop order (called polylogarithms) have a natural operation on them called the 'antipodal map', and in applying this operation to every polylogarithm in the amplitude they recover the form factor. It's intuitive to me that if the mathematical building blocks of your amplitude behave nicely under some operation, you might expect the amplitude to have nice properties under that operation also.So it's clear mathematically why this new relationship exists, and I guess the question is, 'what does it mean physically?'. Dixon et al. comment that they don't know, and I don't really have any thoughts on this either. What I can say is that, over the last 20 or 30 years there have been many different discoveries of duality between seemingly disparate areas of mathematical physics. The first and most famous was the AdS-CFT relationship which relates gravity to gauge theory, and there is lots of work in amplitudes currently on a different relationship which writes gravitational physics as the square of physics in gauge theory. There are many others.So, my experience working in the field has taught me that there are many unexpected dualities and relationships between different physical theories, and that nature is more deeply connected at a fundamental level than is indicated by the initial (apparently disparate) mathematical formulations of those theories.
 “For Dixon, this is a tiny hint at a deeper physical connection between the two amplitudes, and a reason to think something similar might hold in the Standard Model. “But it’s pretty weak,” he said. “It’s, like, secondhand information.””This appears to be a development in a toy model. It might or might not apply to the standard model.Still, at least it isn't a condensed matter experiment being misconstrued as a particle physics breakthrough.
 >> misconstrued as a particle physics breakthroughHey don't throw shade on quasiparticles, they're people too.
 When does a quasiparticle become “particle physics”?As far as I can tell, protons and neutrons are quark-gluon quasiparticles, but we’re happy to call studying protons and neutrons “particle physics”.From Wiki:> complicated system such as a solid behaves as if it contained different weakly interacting particles in vacuumThat is, this description fits protons and neutrons relative to quarks and gluons: very cold quark-gluon regions behave as if protons and neutrons exist.
 Quasiparticles, as usually defined, are embedded in some bulk substrate. So electron holes exist as part of a chunk of metal, for example. Phonons are mechanical vibrations of some bulk matter. They're not made of any particular piece of the bulk matter, like a molecule is made of atoms or a proton is made of quarks; they're emergent phenomena (an electron hole is "made of" not an electron- it's only a hole because of the matter surrounding it).,Protons and neutrons, on the other hand, exist in free space. You can accelerate protons and fire them at the moon (or whatever). You can't do that with phonons or electron holes. They are composite particles (not fundamental) but they aren't quasiparticles.(quarks, weirdly, can't exist in free space. If you try to pull a proton apart you will have to put so much energy into it that you generate more quarks which then immediately couple up with the proton's quarks, possibly in a novel combination producing new composite particles. So you have to study composite particles to study quarks)
 I’m unaware of any “free space” which doesn’t have a mass of quarks and gluons oscillating, so I’m not sure I follow. In fact, the mass of the proton arises from disrupting that typical quark-gluon oscillation.I’m also unsure that’s true, even ignoring that issue:I can shoot an electron hole by moving the chunk of metal it’s situated in, the same way I can shoot a proton by shooting the constituent quarks and gluons.You can’t separate an electron hole from the metal, but you also can’t separate the proton from quarks and gluons.
 Sure, you can take it further and declare that electrons are actually quasiparticles because they're actually excitations of surrounding fields, if you want. I'm not sure what it gets you though- there are real differences between an electron hole and an electron and that's what the quasi- moniker is getting at.
 Sure — but I wasn’t talking about electrons, which as far as we can tell are fundamental particles.I was talking about protons and neutrons.What distinguishes a proton, which gains its mass as a hole in the quark-gluon fuzz permeating the universe, from an electron hole in metal?I think that gets you something:You can understand protons and neutrons as holes in the “crystallized” space that came out of the universe cooling to its current state — that the fuzz of quarks and gluons is something like a “metal” in which protons and neutrons are “holes”.
 It is mathematically useful to think of space as being full of virtual particles including quarks, gluons, and electrons, but there isn't actually a quark-gluon "fuzz" permeating the universe. There was such a physically real fuzz, or more properly a quark-gluon plasma in the extremely early universe (the first few microseconds) which "froze" into hadrons, but this is more akin to water in a cloud condensing into snow.You can certainly formulate the math such that hadrons, or any other particle for that matter, behave like holes in a substrate, indeed this is the basis of quantum field theory. But unlike quasiparticles where that substrate is something in the universe like a block of metal, for regular particles the substrate is the universe.
 I appreciate the time people are taking to walk me through this confusion!- - - - -> You can certainly formulate the math such that hadrons, or any other particle for that matter, behave like holes in a substrate, indeed this is the basis of quantum field theory.Okay — I’m with you here.> But unlike quasiparticles where that substrate is something in the universe like a block of metal, for regular particles the substrate is the universe.Okay — but so what?Why is it different that it’s within a metal than the stuff the universe cooled into? In both cases, we’re discussing an emergent object with new quantum numbers in some substrate:- electron holes are a hole in the fuzz of electrons in metals- protons are a hole in the background fuzz of quarks and gluons in cold aether> It is mathematically useful to think of space as being full of virtual particles including quarks, gluons, and electrons, but there isn't actually a quark-gluon "fuzz" permeating the universe.I’m not sure I follow: proton mass is related to flux tubes suppressing that fuzz in the area of the proton.Empty space has energy, which manifests as that particle fuzz.
 > Why is it different that it’s within a metal than the stuff the universe cooled into?Because that's the point of classification - we want to distinguish between case A and case B. Everything in existence is an emergent property of the universe, but you can't really have any meaningful conversation if the only noun in your vocabulary is "stuff."> protons are a hole in the background fuzz of quarks and gluons in cold aetherNo, there is no cold aether, and protons are not holes in anything. They are composite particles. To the degree that their constituent parts can be described as excitations of quantum fields, so can they, but they are no less real.> I’m not sure I follow: proton mass is related to flux tubes suppressing that fuzz in the area of the proton. Empty space has energy, which manifests as that particle fuzz.Proton mass comes from the confinement of its constituent parts. Only about 9% of the proton's mass is a result of the quarks it contains interacting with the quark field, which again is not literally a "fuzz" of discrete particles. Flux tubes are just shapes of magnetic fields. Protons are not just their mass, they have many properties.
 > Because that's the point of classification - we want to distinguish between case A and case B.Right — but why?What is useful about that classification if the two types of particles emerge the same way, as holes within a substrate?What do we gain in “composite” vs “quasi”?> No, there is no cold aether, and protons are not holes in anything.There is — we can measure vacuum energy as existing. We can also measure spacetime as existing, directly. We know there’s an aether.> Proton mass comes from the confinement of its constituent parts. […] Flux tubes are just shapes of magnetic fields.Flux tubes are the region where the connection between quarks suppresses the fuzz from vacuum energy. That’s where the non-quark mass comes from. (As you call it, confinement.)
 cwillu on Aug 1, 2022 “I identify as Higgs”
 Queue "yo mamma" joke.
 Can a physicist comment on the impact of this to the work by Nima Arkani-Hamed et al on the Amplituhedron (https://en.wikipedia.org/wiki/Amplituhedron)?
 Great question! This result is in the same area as the work on the amplituhedron. It's not immediately obvious to me that they are directly related, and I didn't see a reference to any amplituhedron papers in the citations of this paper. Dixon et al. have essentially discovered a new symmetry which the scattering amplitudes (of N=4 Super Yang-Mills) must obey, and new symmetries generally lead to deeper understanding of the theory. This symmetry should then also be somehow encoded in the amplituhedron description of the theory; I imagine people would look into this in the future.
 Cool, thanks!
 Very pleased to see news like this -- observations of this sort are where the big theoretical advances sometimes emerge.
 Indeed, this is the sort of stupefying discovery that hints at something deeper
 As a complete layman to all this, after reading the article it feels like more proof we are in a simulation.
 No one said the computer running the simulation used digital logic. If we could encode system dynamics into "energy+matter", you could call it an analog computer without much dissent. People do this in miniature: in fact, the first computers were entirely mechanical!If we accept this reasoning to imply that the universe is a giant computer, we only have the unknowable left: was it purposefully (intelligently) designed or did it spontaneously arise? As far as I can tell, the probability of either outcome is equal according to our limited experiences, and we will never have access to a definitive answer one way or the other.
 The first sentence of the abstract is a helluva mouthful: "We observe that the three-gluon form factor of the chiral part of the stress-tensor multiplet in planar N = 4 super-Yang-Mills theory is dual to the six-gluon MHV amplitude on its parity preserving surface."For any of you paper-writers out there, some advice: the abstract shouldn't contain undefined acronyms, and should be readable by anyone who you would expect to be at a conference where you present your paper. Exchanging a bit of the technical correctness for readability is the entire point of the abstract, and the source paper's abstract could really use some work.
 I have a PhD in this field. I understood everything in the title straight off, apart from the context of which 'parity preserving surface' they were talking about (which was fine). I think this was a concise and well-written abstract, which was easily readable within the target audience. There are very many different concepts that you have to build on and learn to be able to write something like this paper. It's unfortunate that abstracts in the field are not more accessible to a wider range of people, but I don't see an easy solution to that. If you went to the yearly conference on this field and started asking people to define 'MHV' in their abstracts I don't think you'd get many converts. This acronym is cryptic, unintuitive and completely ubiquitous and ingrained in the literature.
 Every writing course I attended in grad school told us that the first sentence of the abstract should be understandable by any scientist, the second sentence should be comprehensible to a scientist in a related field. After that, you address the specialists community. However, it seems that more and more people are skipping the first and second sentences and jumping directly to talking to specialists. I have mixed feelings about that. On one hand, it's nice to skip the generalities but at the same time some times its nice to have some context.
 ISL on Aug 1, 2022 That sentence is very readable for anyone who would be at a relevant conference.As an experimental physicist, everything but the MHV (maximally helicity violating) was readable/comprehensible to me. Theorists in the field will read that as easily as breathing.
 Context is king. It's the abstract of a paper, not simple Wikipedia. Everyone reading the paper itself is expected to either know or make the effort the become knowledgeable enough to understand. In any other context, I'd agree but we're taking about a paper on theoretical physics on a potential new interaction lol
 I think this would be a great task (and challenge) for AI: Make abstracts understandable to a wider audience.
 I take it as a sign that I would have to understand a bunch of prerequisites before I could understand the content of this paper, its implications, and the possibilities it raises.
 It's a paper, not an introductory textbook...
 Precisely...
 In a field like this, the abstract is more about conciseness than accessibility. It's for a fellow specialist skimming through dozens of papers looking for something interesting.
 Do you maybe have a concrete suggestion for how you would rewrite it? Which parts do you think are not understandable for anyone attending, say, the annual Strings conference?
 > and should be readable by anyone who you would expect to be at a conference where you present your paper.Yes, I guess this is pretty much the case here. There's no much way to simplify thisConciseness is important in abstracts and it beats general simplifications.
 The abstract looks perfectly fine - it is written so anyone working in this area would know what the paper is going to do. Writing it at a basic level would either require the abstract to be absurdly long or to leave unspecified what the paper contains.I'm not in particle physics, but have read textbooks and papers in many areas of theoretical physics, and I understand enough of this abstract to know what it's going to do.If MHV is an acronym you think should be expanded, it is not. There's well over 500,000 hits on google for MHV physics, and only 2900 hits for the expanded term. It is a widely used term in the field.
 They're not writing for you.

Applications are open for YC Winter 2024

Search: