8 ms·
Making math more Lego-like
- kmill 10y agoRelated work: * Penrose graphical notation. https://en.wikipedia.org/wiki/Penrose_graphical_notation https://en.wikipedia.org/wiki/Penrose_graphical_notation * Trace diagrams. https://en.wikipedia.org/wiki/Trace_diagram https://en.wikipedia.org/wiki/Trace_diagram * Trace diagrams for Hopf algebras. http://tmp.esoteri.casa/hopf.pdf http://tmp.esoteri.casa/hopf.pdf * Frobenius algebras. http://math.ucr.edu/home/baez/week299.html http://math.ucr.edu/home/baez/week299.html * Temperley-Lieb algebras. https://en.wikipedia.org/wiki/Temperley%E2%80%93Lieb_algebra https://en.wikipedia.org/wiki/Temperley%E2%80%93Lieb_algebra * Planar algebras. https://math.berkeley.edu/~vfr/VANDERBILT/pl21.pdf https://math.berkeley.edu/~vfr/VANDERBILT/pl21.pdf * String diagrams for 2-categories. https://en.wikipedia.org/wiki/String_diagram https://en.wikipedia.org/wiki/String_diagram * Diagrammatic algebras in general. http://mathoverflow.net/questions/168888/who-invented-diagrammatic-algebra http://mathoverflow.net/questions/168888/who-invented-diagra... * Begriffsschrift. https://i.ytimg.com/vi/KFYvF3t1xUM/hqdefault.jpg https://i.ytimg.com/vi/KFYvF3t1xUM/hqdefault.jpg * Existential graphs. https://en.wikipedia.org/wiki/Existential_graph https://en.wikipedia.org/wiki/Existential_graph * Laws of Form by Spencer-Brown. https://en.wikipedia.org/wiki/Laws_of_Form https://en.wikipedia.org/wiki/Laws_of_Form * Opetopes. http://mat.uab.es/~kock/cat/zoom/scan.jpg http://mat.uab.es/~kock/cat/zoom/scan.jpg
- closed 10y agowell, there goes my Friday night. Cheers :)
- pizza 10y agoalso amplituhedra for computing particle interactions
- ramblenode 10y agoBob Coecke's "Kindergarten Quantum Mechanics": http://www.cs.ox.ac.uk/bob.coecke/VaxjoProc.pdf http://www.cs.ox.ac.uk/bob.coecke/VaxjoProc.pdf
- laingc 10y agoThere is precisely one person who can be productive with Penrose notation, and that's Roger Penrose. You should see him do it, though - very impressive.
- OscarCunningham 10y agoThis isn't true. I've personally found proofs in the diagrammatic notation that I wouldn't have found without it.
- sfgysfdyu 10y agoalso similar to work by Louis H. Kauffman in representing QM with knots & braids https://arxiv.org/pdf/0903.1376.pdf https://arxiv.org/pdf/0903.1376.pdf
- mathgenius 10y agoGlobular: https://ncatlab.org/nlab/show/Globular https://ncatlab.org/nlab/show/Globular
- trustthispost 10y agoThere's a history from the introduction of Tensor networks by Penrose to the 3D tensor networks, the quon language: https://dl.dropboxusercontent.com/u/78485948/PNAS-2017-Biamonte-1700736114.pdf https://dl.dropboxusercontent.com/u/78485948/PNAS-2017-Biamo...
- trustthispost 10y agoThis gives more of a physical development, and frames the context of the lego-like language in the cross-disciplinary field of quantum information
- egonschiele 10y agoThis sounds incredibly cool, but how can I see an example of the images? A simple search for "Quon examples" yielded nothing.
- dongslol 10y agoThe paper has a few: http://www.pnas.org/content/early/2017/02/03/1621345114.full http://www.pnas.org/content/early/2017/02/03/1621345114.full
- deepnet 10y agoQuons: A 3D Language for Quantum Information Liu, Wozniakowski, & Jaffe https://www.arxiv.org/abs/1612.02630 https://www.arxiv.org/abs/1612.02630 Holographic Software for Quantum Networks Jaffe, Liu, & Wozniakowski https://arxiv.org/abs/1605.00127 https://arxiv.org/abs/1605.00127
- trustthispost 10y agoHere, you'll find discussion on the development of the picture-language and some application, i.e., topological design of protocols (by Jaffe, Liu, and Wozniakowski): https://arxiv.org/abs/1611.06447 https://arxiv.org/abs/1611.06447
- minimaxir 10y agoIn deep learning, layer manipulation in Keras is often compared to Legos, as you can mix-and-match connections between layers and stack them pretty easily with minimal code adjustments. It makes iteration/experimentation extremely easy.
- nathancahill 10y agoI generally keep memes off of HN*, but this one always cracks me up: http://i.imgur.com/GYdvtAW.jpg http://i.imgur.com/GYdvtAW.jpg [so I'll delete it if it gets downvoted]
- chris_chan_ 10y agoA wise man once told me visual image speak louder than words. Yes, this sounds incredibly awesome.
- ballenf 10y agoIsn't it ironic that the article had not a single picture of said picture models? (You can't count the rough chalkboard sketches as 3D model, imo.) The actual linked paper does, of course.
- aviggiano 10y agoI was also impressed by the fact that the news title talks about 3D lego models but has no images whatsoever
- mehaveaccount 10y agoUniversity magazines are a terrible place to learn about new research because they are effectively trying to sell the research at their university as great whether it is really great or not. Looking at the arxiv link they linked I was underwhelmed compared to the title. It is not a new lego-like language for "mathematics", it's some definitions in a very particular subfield in mathematics. An accomplishment within that subfield to be sure, but not anything relevant to the wider math community and certainly not Hacker News.
- Amygaz 10y agoMy personal experience with Harvard is they are the best at over-exaggeration. Also very good at branding "been done before, but not at Harvard" research as brand new and revolutionary. Others do it too, but HU masters this.
- billforsternz 10y agoI would have expected "making math more legos like" (US English) or "making maths more lego like" (British English). Instead we have a hybrid version.
- psquid 10y agoit's US english, they're referring to Lego the product as a whole, rather than individual Lego pieces (the meaning ends up basically the same, but the distinction does change the grammar)
- fjdlwlv 10y agoLego is not a product. LEGO is a trademark for a brick system.
- billforsternz 10y agoI see. My impression is that in US film and television "legos" is used almost exclusively. But it's not like I've been taking notes or anything, and I could well be wrong.
- trustthispost 10y agoIf you want an intermediary between the linked Lego story and the quon paper in PNAS, then check out the commentary (in PNAS): https://dl.dropboxusercontent.com/u/78485948/PNAS-2017-Biamonte-1700736114.pdf https://dl.dropboxusercontent.com/u/78485948/PNAS-2017-Biamo...
- elchief 10y ago> It turns out one picture is worth 1,000 symbols Says article with ~750 words and one useless picture
- gravypod 10y agoWhy do this rather then define an abstract, human-evaluatable, programming language? Give it some S-expr syntax and make more things implicit and you're golden. Making strange squigles on a page doesn't help. Making strange greek symbols on the page doesn't help. Writing functions, "unit tests", building on abstractions, and creating "libraries" do help. We all know this because it works well in CS. Why would this not work well in physics? Ditch the alphabet soup (and yes, even the 5th centry Greek alphabet soupe). Abstract common ideas into "functions". Write human-provable "test-cases". Add documentation. Use logically derived variable names (even if they're 5-10 characters). Formalize the definitions and syntax of this language and it's abilities. You'd be golden. Phi shouldn't mean 30 things depending on the field you're in. Am I wrong? Is this not possible? Is there a reason as to why math and logic can't be expressed as an abstract programing language? Is there a reason why it wouldn't be better to have a completely standard way as to how to define all of your algorithms and logical assertions? I could be wrong but no one has provided information as to why what we do is better. Nor does this look much better (what kmill has linked).
- doubleunplussed 10y agoWhen you write a function, you still need to give the arguments names. When you write an algorithm and want to share it with someone, you still need to give the variables names. That's all the greek symbols are - and mathematical notation is an abstract, human evaluateable programming language. Common ideas are abstracted into functions, like sine, cosine - min, max, even weird things like Bessel functions, Legendre polynomials, these are all written as function(arguments) in maths. Maths has indexing, a bunch of operators, and what the operators do often depends on what type of objects they're applying to. Python's list and set comprehensions etc actually come from a real mathematical notation for defining sets. All those capital sigmas and capital pi's are reduction operators - sums and products. There isn't an assignment operator - you're expected to give each variable a unique name if you have a multi-stage process, so you only ever make statements of equality rather than assignment. Sure, it's an ambiguous language sometimes, but it sounds like your complaint comes down to standardising it more and using longer variable names.
- 10y ago
- andrepd 10y agoWith all due respect, you have no idea what you're talking about, you aren't making any sense, and you're clueless about what mathematics is about when you call for ideas to be abstracted into "functions", and complain about naming conventions and so on. In your eagerness to apply what you've learned about software development to a completely different field you are grossly and misunderstanding what it is all about. Your posts reveal a fundamental misunderstanding of what mathematics (and physics) is about. Tip: it's not about algorithms. So don't presume you can apply techniques from software development to mathematics, just as it doesn't make sense to apply tools from physics to computer programming.
- gravypod 10y ago> With all due respect, you have no idea what you're talking about, you aren't making any sense, and you're clueless about what mathematics is about when you call for ideas to be abstracted into "functions", and complain about naming conventions and so on. In your eagerness to apply what you've learned about software development to a completely different field you are grossly and misunderstanding what it is all about. You're right, I don't know much about the fields of mathematics and physics. > posts reveal a fundamental misunderstanding of what mathematics (and physics) is about. Tip: it's not about algorithms. So don't presume you can apply techniques from software development to mathematics, just as it doesn't make sense to apply tools from physics to computer programming. Actually quite the opposite. I do expect programming principles and tools to be able to aid in expressing ideas that can be thought about as a language. This is just visible with problems like unstandardized common variables. The only physics heavy book I've read some of recently is Kenneth Davies Ionosphereic Radio Propagation because it relates to my work. In it, there is a few pages dedicated to just defining what all of these funny greek symbols mean because they are used in pretty much every math heavy disiplin but each field takes them to mean their own thing. Even still with this quick referense there are large portions of the book that remain a mystery because they simply require understanding of this single field of physics common variable names and meanings to read. The chapter on VLF propegation (just after the diagram of these waves being propegated by the Earth's field lines) refers to X sub something (it's been a few days) and it relates that approximately to the plasma frequency at a region of space but it means something different. I have no idea what that variable is without finding a plasma physicist and asking them directly what that specific niche variable name means for this specific field under this specific topic relating to this one type of equation. Even with this simple example, I can show that it would be better to have a discriptive variable name rather then "X sub something". This is just one example from this week where this trend has bit me in the ass. If it's so easy to talk about this single situation where it's the case that a programming idea of descriptive variable names, then why don't other programming concepts fit in math? Can you prove why this is the case?
- mattxxx 10y agoCool.
- mrcactu5 10y agoall these wonderful graphic languages exist. now it's time to put them into practice!