Fixed Points
Video Statistics and Information
Channel: Vsauce
Views: 6,463,649
Rating: 4.9325085 out of 5
Keywords: fixed points, math, education, vsauce, michael stevens, topology, shapes, brouwer's fixed point theorem, borsuk-ulam, temperature, geometry, science, infinity, numbers, functions, learn
Id: csInNn6pfT4
Channel Id: undefined
Length: 16min 25sec (985 seconds)
Published: Wed Sep 28 2016
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.
I've always felt like fixed points are an underappreciated concept in mathematics. ex is a fixed point of the derivative. Eigenvectors are sorta fixed points (for Markov matrices they are exactly). Newton iteration heads towards a fixed point. The golden mean is a fixed point of f(x) = 1 + 1/x. The normal distribution is a fixed point of the fourier transform. Any discussion of chaos theory usually starts with a section on fixed points. It goes on forever. Really, I think the concept deserves an entire book for itself.
What about the filthy Frank collabs Micheal?
At the end he claims that, if you start with a number, count the letters of its name in English, and apply the same process to the result iteratively, you always ends up with four. His argument is that four is the only fixed point of that operation. But he doesn't rule the possibility of loops or going to infinity, right? I am willing to believe the claim is in fact true because if you start with a huge number or a non natural number, the process quickly turns the number into a small natural number, but he doesn't mention this, or does he?
Well, in the coffee example several assumptions need to be mentioned, like you are only considering two time instants, the stirring is continuous and every point in the surface stays in the surface, etc.
Also, I was expecting some mention of the Hairy Ball Theorem.
vSauce just needs to go ahead and get a math degree already.
Is he secretly working up to a video on the incompleteness theorems? First ordinals, now fixed points, I believe he had a video where he went into more axiomatic arithmetic some, all he'd need to do now would be to cover basic proof stuff in an interesting way (possibly a video on induction? Sounds like his style) then he could do one on the foundational crisis and russels paradox, and then he could easily make a long one where he sketched the whole proof out well enough for his audience to understand
eye-DEM-poh-tent?? No, no, I don't think so. EYE-dem-POH-tent, you scoundrel.
At 7:21 isn't it a mistake? The number of right digits of the approximation will not double with each iteration. I'd say that for each iteration the approximation will get "half a digit more rightο»Ώ". The example from wiki.
EDIT: I just calculated x_4 and x_5 from the wiki example and they were accurate to the 8th and 18th significant figures so I guess I was wrong and the number of right digits of the approximation is indeed doubling with each iteration of the babylonian method which I didn't believe some ancient technique would do.
Is his statement about the coffee accurate? Wouldn't the stirring of the water molecules essentially "tear" them away from each other as they get mixed up, since water is a liquid, rather than a solid?