Post

Conversation

Lucas Beyer (bl16)
@giffmana
I'm starting to think mathematics were not beautiful because it "describes the beauty of nature" but rather because human mathematicians only went after beautiful but suboptimal descriptions lol
Image
Quote
Quantіan
@quantian1
You can do Fourier transforms in O(N log(N)^0.9999999999999) time. Fuck this gay chud universe
Image
89.2K
Views
View quotes
David Watson 🥑
Post your reply

George Pearse
@Georgehwp1
Feels like such a strong analogy to compilers and programming languages. Surely surely you can find ways to express the ideas that retain the beauty. Can’t pretend I know maths well enough for this to be anything beyond shooting from the hip
886
Noam Shalev
@noamshh
"My work has always tried to unite the True with the Beautiful and when I had to choose one or the other, I usually chose the Beautiful." ~ Hermann Weyl
3.1K
Nicholas Bardy
@NicholasBardy
But the ugly stuff is still made of the beautiful simple building blocks. It the primitives that compose that are the beauty of nature.
260
Alan
@AlanIsCurious
“The discovery [proof that sqrt(2) is irrational] posed a problem that was central to Greek mathematics. The Pythagoreans had, up to this point, identified number with geometry. But the existence of incommensurable ratios shattered this identification.” A tale as old as time!
591
Jatin Garg
@jatingargiitk
which areas of math stayed beautiful but suboptimal the longest? numerical analysis seems like one where ugly-but-fast methods won.
2.4K
tuomo
@7uomoki
or the example here was urgly human construct to begin with! i reger to the squares within a square thing
1.6K
AlexWD
@alexwd
I wouldn't say that mathematicians go after suboptimal things, but it is potentially true that mathematics has been biased towards solving problems that have beautiful solutions.
655
vim
@agent_cto
This may be the main reason for not having made many breakthroughs in physics as well. From now on I will let Claude write any type of ugly code(to me) that it thinks that is great
1.1K
Aditya Timmaraju
@tadityasrinivas
It's still possible that there are "beautiful", yet to be discovered special functions/abstractions that elegantly describe and underpin the new results. It has, in fact, happened several times before in math that ugly results come first.
1.1K
Amo
@hydraulicspood
We should give up on math and physics being beautiful and start looking for the garbage-looking numbers and formulas like raccoons
1.8K
TSMN
@tsmndev
Why is this about a universe thing? The most optiomal algorithm AI and we found for now is looking like this. There will probably be more optimal algorithms found in the future. We can not make a claim for now
712
Christian Acevedo
@Nik_tes
Not true in all cases, but many times mathematicians ended up having, indeed, beautiful suboptimal descriptions of reality. We were ok with that in other times in humanity. Right now we are in an age lf optimization over everything, even if we do not understand how it optimizes
191
Ben
@Ben18262517
Perhaps the weird solutions are beautiful in whatever space the LLMs reason in. If you already encode numbers as helices or whatever, who knows where you find non-human beauty.
1K
Émile Amajar
@emile_amajar
Relevant. Ugly problems have ugly solutions.
Quote
Émile Amajar
@emile_amajar
Replying to @dcolascione
It's because THE PROBLEM is ugly: packing 17 unit squares within a square of side 4.67553
25
gatsby
@goatsbee
Exactly, as much as I hate this,kinda happy that grifters(who often couch lack of technical leg work cause beauty/intuition etc) get burned cause of this.
857
gautam
@grayap9
I still believe its because the models are "brute forcing" it. Once humans get a chance to look at it more closely a lot of these results will get nicer
207
wkdnd
@Chawkkkk
Interesting idea! No doubt aesthetics may have played a large part in the development of mathematics. But maybe we like aesthetics because it’s harmonious with the way the universe fundamentally is?
371
Guillaume Ausset - @ausset.me
@aussetg
Not gonna lie, if a bound, equality or whatever, I’m trying to derive doesn’t end up with either 0 or some neat expression involving pi, exp and sqrt(n) I basically assume I screwed up somewhere
238