« first day (2417 days earlier)      last day (2605 days later) » 

12:10 PM
@Secret You can make a curve as curvy as you want on the plane.
 
Is curvature of a planar curve being uniform a well defined term, cause it seems to imply constant curvature to me?
 
It's not a thing
Hi @AkivaWeinberger
 
I see
 
Hi
I wonder why people write $\exp(\text{really long formula})$ rather than $e\text^(\text{really long formula})$
Oh, the ^ sign looks hideous in that font
 
it looks nice
the stuff above the e gets like really small if it's something huge
$e^{e^{n\log n+O(n^{1/2})}}$
now you can't tell whether that's a little-o or a big-o
 
12:18 PM
Also, writing $2^n\text{ where }n=\text{really long formula}$, I think, rather than $2\text^(\text{really long formula})$
 
^ is ok but horrible
 
Hi @Alessandro
 
@BalarkaSen For base $e$, you can use $\exp$
 
Hi @AlessandroCodenotti
 
12:20 PM
Other bases are more annoying
 
@AkivaWeinberger That's a good point. I don't know.
Ask an analytic number theorist.
 
Hm, I wonder if people ever use ${\Large o}(x)$
It's literally a big o
(Kidding)
 
I wrote a complicated pro-existentialist surreal neo-Dadaist analysis of Octavio Paz. Not sure how the examiner would take it but I really wanted to pull that stunt.
 
Had to scratch on my computer screen to check if that's really a genuine tick above the O or a spot on my screen.
 
12:31 PM
They're eyebrows
ò_ó
 
o you guys onto philosophy currently eh?
2
Q: How are beliefs restricted by an objective reality, and do unreal beliefs can only cause real impacts indirectly?

SecretEver since reading this NewScientist issue on how even if for some beliefs that despite being based on falsehoods, can have actual impacts to the world via cultural influences, I started to wonder about the nature of belief, and by extension, unreal entities or concepts that have real impacts. C...

 
magic of ascii
@secret nah
 
ok
(anyway, the two question I asked there turns out to be way TOO BROAD that even splitting them into two separate question is still TOO BROAD)
Joke: start to wonder whether my question has a cardinality of at least $\aleph_{{\omega}^{\omega}}$
(in a few minutes, the stuff I derived earlier this morning will be continued, hoepfully this time they don't look so ugly)
 
$\rm Back^{Future}$
 
meh
 
12:45 PM
Hey guys I have complex analysis class at university and we currently look at integrals, unfortunately we don't proof the theorems, since this is more like a crash course for electrical engineers.

I read through some papers about the proofs for the Cauchy-Integral theorem, but it's so much stuff and there seem to be a vast number of interpretations of the Cauchy theorem.
Oups, not finished yet
When we want to calculate the circle integral of a holomorphic function $f$, but there is a singularity inside the circle, why do we have to use the cauchy fromula? Since when we evaluate $f$ in a circle around the singularity we never touch that "bad" point actually no?
Same with wholes in sets, e.g. unconnected sets...is this due to missing homotopy of our circle function?
 
@Felix.C There may be bad points inside the circle, is the point. Eg, $f(z) = 1/z$ is a holomorphic function which has no singularities on the unit circle $|z| = 1$. But $\displaystyle{ \int_{|z| = 1} \frac{dz}{z} }$ is not zero.
 
1:01 PM
@Felix.C It's because holomorphic functions on $D\subseteq\Bbb C$ are not guaranteed to have global antiderivatives. If $D$ is simply connected, this becomes true, but that's kind of a miracle
For functions with antiderivatives, the fact that their integrals on closed loops equals zero follows from the Fundamental Theorem of Calculus.
 
You don't *have* to use the Cauchy formula, in the sense that you can just use
$$\int_{|z|=1} f(z) dz = \int_0^{2\pi} f(e^{it})ie^{it}dz$$
 
What's true is that every holomorphic function has local antiderivatives around any point. If $D$ is simply connected, you can patch it all up to a global antiderivative. If it has singularities (like if it's $\Bbb C^*$) then everything can't be patched up.
 
Yeah. 'Cause, locally, they look like linear functions (because they're differentiable), which have antiderivatives
 
1:16 PM
Anybody know about crystals in rep theory?
 
crystals?
As in, crystallographic groups?
 
Nono Crystals as in Kashiwara Crystals of kac-moody algebras
 
then no :p
 
vO-ov
 
I didn't expect anybody to, just seeing if there is a hidden expert
 
1:20 PM
So if I understand this right, it boils down to the problem that holomorphic functions $f:D \subseteq \mathbb{C} \mapsto \mathbb{C}$ with $D$ not being simply connected, aren't guaranteed to have a global antiderivative $F:D \mapsto \mathbb{C}$, therefor we can't use the fundamental theorem of differential calculus on them, but rather have to watch local antiderivates...such that the integral won't be necessarily 0.

Then is there a theorem about the situation where you have a whole in your set, and you want to calculate the integral around it?
 
if anyone can it's Tobias Kildetoft. But he hasn't been around for a while.
 
I don't think Tobias does that stuff
 
He's big on representation theory
 
I spoke to him a bit back I don't think hes into modern stuff yet
oh shit wait I am confusing people
hmm
 
I think what he does is pretty modern :P
 
1:25 PM
I think your chances of finding an expert in such a narrow branch are very slim
 
I know he does lots of group theory stuff
but I think he is mostly interested in non-lie groups
I could be very wrong
I shall ask him when he is around
 
You seem to be assuming a lot about a professional representation theorist.
 
I work on things with a Lie structure quite a lot, and I'd never heard of those crystals...
 
@BalarkaSen only from what I gathered from talking to him
to be honest I don't think I can seriously say much because we never had a serious conversation about rep theory
Crystals arise natrually in combinatorial descriptions of reps of lie algebras
 
From Tobias' publication list, there doesn't seem to be anything related to semisimple Lie's or quantum groups
 
1:27 PM
As you say it is niche
But still a lot of mathematicians publication lists have pretty much nothing to do with what they actually know
Except for mine
mine is pretty much accurate
I have no publications as I know nothing :P
 
Yes and no... many mathematicians know much more than their publication list tells you
 
The best way to find out is to ask or force yourself into their office
 
But they rarely know other things on a level of being able to do research on it
 
It may be true that Tobias can not answer your question. That does not mean he doesn't do "modern stuff". He's a serious rep theorist. That's all I am saying
 
@BalarkaSen I misspoke I was thinking of somebody else
my modern stuff comments don't apply to Tobias
 
1:30 PM
(He's probably your best bet in this chat if you want your question answered anyway)
 
I'll make sure to ping him later
@BalarkaSen What have you been getting upto these days?
 
exams
in math, still learning foliations
 
I read a bit more about foliations, they look cool
 
they're good
 
I've been working a bit more through Hatcher, homological algebra really doesn't sit unless you understand the topology.
 
1:35 PM
true
not for me at least
 
I like the interplay between cohomology and homotopy
 
right. my favorite slogan is "cohomology is dual to homotopy"
 
i have seen similar constructions in algebraic geometry as well (with sheaves) but I haven't been concerned with homotopy much there
 
Hmmm... If $\exp \sum_{n=1}^\infty \frac{a_n}{n}z^n$ and $\exp \sum_{n=1}^\infty \frac{b_n}{n}z^n$ are both rational functions, is there a reason to assume that $\exp \sum_{n=1}^\infty \frac{a_nb_n}{n}z^n$ will be rational?
 
@SteamyRoot I don't think you can
 
1:38 PM
which constructions do you have in mind?
not sure if we have the same things in mind.
 
Hmm, okay. I guess it's worth investigating then.
 
@BalarkaSen Brown had written an article on it. I read it a while back let me see if I can find it
 
what i was thinking of was [X, K(G, n)] = H^n(X; G)
 
Yeah I only looked at the shiny results but here it is
@BalarkaSen yes its an extension of that of sorts but for cohomology of sheaves
 
that's mildly interesting
 
1:42 PM
@BalarkaSen Yeah for now its a mild curiosity. I don't think I will study anything serious enough to warrant pulling it out for the time being.
 
do I know what classifying space of a sheaf means? shrug
 
@Felix.C I think it's called the residue theorem when that hole is a point
(or a finite set of points)
 
@BalarkaSen I think it arises from Grothendiek touching topology
 
Oh I guess this has something to do with Brown's representability theorem
(you replace the Eilenberg Maclane space by a spectra for a generalized cohomology theory)
Not sure how the story goes in the category of sheaves, which is a lot more rigid but yea
 
1:46 PM
@AliCaglayan Everything arises from Grothendieck
 
They talk about the classifying space of O(n) modelled using Grassmanians
 
The K-theory groups are classified by maps to BSO(n)
 
huh interesting
 
which I guess is quite close; K-theory classifies vector bundles which are in particular sheaves
 
hmm this does all G-bundles
 
1:47 PM
something like that is happening for varying fibers
@AliCaglayan yes, for G-bundles it's pretty easy. They are classified upto isomorphism by maps to BG
I am just not sure how to do it for varying fibers
 
hmm interesting
has applications to higher van kampen
Lots of reading for the future
 
I probably won't read it
#toohard4me
 
Only the cover
the rest depends on how much time you are willing to spend
 
2:32 PM
Can I ask a question?
Anyone there?
 
2:57 PM
Trying to listen to a talk when you're not sufficiently rested => spend most of the talk straining to stay awake
 
Hi guys, I will really appreciate if anyone could please help me look at this question on vertically homotopic question math.stackexchange.com/questions/2190070/question-on-homotopy
Thanks for helping
 
@Semiclassical so true
 
3:28 PM
corollary: the more energy you need to use to follow the talk, the more you'll be straining to stay awake
 
It helps to furiously try and copy the blackboard, even if there's no point in doing so
 
doesnt work so well for PowerPoint/Beamer presentations
 
Ah...
Rarely if ever have those :P
 
Eh, that's what all these workshop talks have been
I guess the corollary being: it's harder to stay awake during applied-math talks :)
3
 
3:57 PM
Started a Discrete Math assignment 9 am in the morning. Spent all day working on it with minimal breaks until 4 am. I managed to wake up after a few hours of sleep, made my way to Uni, and walked into class only to hear my professor announce that the assignment is not due Monday, minutes before the deadline. Terrrrrrrrrrrrrrrrrific.
I was sleep deprived and stress-induced for nothing :)
 
4:18 PM
my problem is i frequently sit up front
 
4:41 PM
Hey guys!
 
[The Integral Exercise]
Recall that earlier in the afternoon we have the following:

$\int a(n)x^n dn=\sum_{k=0}^{\infty} (-1)^{k-1}a^{(k)}(n)(x^n)^{{-(k+1)}}$

Now consider $x=e^u$. Thus we now have:

$\int a(n)e^{nu} dn=\sum_{k=0}^{\infty} (-1)^{k-1}a^{(k)}(n)(e^{nu})^{{-(k+1)}}$

Now let $w=e^{nu}$, hence $\int w dn=u^{-1}e^{nu}=u^{-1}w$, $w^{(-k)}=u^{-k}w$. Therefore:

$\int a(n)e^{nu} dn=\sum_{k=0}^{\infty} (-1)^{k-1}a^{(k)}(n)(e^{nu})^{{-(k+1)}}=\sum_{k=0}^{\infty} (-1)^{k-1}a^{(k)}(n)u^{-(k+1)}e^{nu}$
Now for this limit:

$\lim_{R\to \infty} (-1)^R\int a^{(R)}(n) \int ^{(R)}e^{nu}d^{R}ndn=\lim_{R\to \infty} (-1)^R\int a^{(R)}(n) u^{-R}e^{nu}dn$

Assuming $\lim_{R\to \infty} a^{(R)}(n)$ exists and is bound. Then $\lim_{R\to \infty}\int a^{(R)}(n) u^{-R}e^{nu}dn=0$ and hence the terms in the series goes to zero as $R \to \infty$
Therefore

$$\int a(n)e^{nu} dn=\sum_{k=0}^{\infty}\frac{(-1)^{k-1}a^{(k)}(n)e^{nu}}{u^{k+1}}$$
 
4:57 PM
@MikeMiller I have a quick question. So if we have $p : E \rightarrow B$ where B is path connected. Then I am trying to prove that the fibers are actually all homotopy equivalent.
I proved that we get vertical homotopy equivalence
as before yesterday
I considered the following diagram
Let $b_1$ and $b_2$ be points in B let $\phi : I \rightarrow B$ is a path between $b_1$ and $b_2$.
consider the following diagram
$p^{-1}(b_1)\times \{0\} \rightarrow E$
on the bottom square we have
$p^{-1}(\{b_1\}) \times I \rightarrow I \rightarrow B$ where the second map is $\phi$
now I am trying to prove we get homotopy equivalence given this square
 
5:13 PM
Now use $e^x=\sum_{n=0}^{\infty}\frac{x^n}{n!}$ we get:

$\sum_{k=0}^{\infty}\frac{(-1)^{k-1}a^{(k)}(n)e^{nx}}{x^{k+1}}=\sum_{k=0}^{\infty}\frac{(-1)^{k-1}a^{(k)}(n)\sum_{m=0}^{\infty}\frac{x^m}{m!}}{x^{k+1}}=\sum_{k=0}^{\infty}\sum_{m=0}^{\infty}\frac{(-1)^{k-1}a^{(k)}(n)x^{m-k-1}}{m!}=\sum_{k,m=0}^{\infty}\frac{(-1)^{k-1}a^{(k)}(n)}{m!}x^{m-k-1}$
which is some kind of alternating laurent series
 
@Adeek What is $p$?
 
5:29 PM
Hi chat
 
Hey @Astyx
 
a fibration @BalarkaSen
 
What's up ?
 
Not much, just getting some post-finals relaxation
 
@MikeMiller Ah
 
5:33 PM
How about you?
 
Starting to get stressed about my exams
 
Lmao, good luck
 
Yea. Exams do that. They come with stress at no additional charge.
 
Thanks
What were your finals about ? @Daminark
 
The only test was in analysis
 
5:36 PM
What kind ?
 
This was functional analysis and a bit of manifolds
 
Are you in Uni ?
 
Yup
Wait huh I was always under the impression that you were as well
 
What's it like ?
 
I definitely like it here, it's fun
A good deal more work than in high school for sure, but it's definitely more interesting so it all balances out well
 
5:39 PM
I'm not exactly in Uni, I take classes préparatoires which is preparation for competitive exams for engineering schools
But I have a double cursus with Uni (or at least I would have had if I had sent the registration form in time)
 
@Daminark What do you think of those who go around saying "Universities are a waste of money and time"
 
I wouldn't really agree with that assessment. It's definitely not like before when going to college just kinda meant you were set for life, so for a lot of people, especially if you don't major in those subjects deemed "practical", it'll be stacking on debt without the guarantee of career security
But it's still better to have gone to college than to not, from a purely practical standpoint, and I find education to be an end in itself
(Also, I have financial aid which really helps)
Admittedly, it can be context dependent, but at least for me personally, this timeline is far better to one where I don't go to college
 
A very reasonable way of viewing it. I find myself in a similar situation too. Graduated from college, and currently in Uni. I find that University gives you a "reason" to keep moving forward. Sort of like a momentum, and a purpose.
Also, financial aid is nice and all, until you start paying it off :P
 
Ah, these are actually grants, luckily, so there's no need for that worry
And I just wanna be clear about terminology because I suspect that it differs between us
For me, college is often used synonymously with university for undergrad stuff
 
Ah. So much confusion when it comes to these two terms. It seems like everyone I talk to, has a different meaning for College and University.
 
5:52 PM
Im pretty sure that in the US the term college is pretty much applied to any bachelor's degree granting institution
whereas universities can also offer more advanced postgrad degrees
 
Ah. I'm in Canada and Colleges seem less significant than Universities.
Universities here have a higher standard, and offer a wider range of advanced degrees.
 
Sometimes down here a university often "contains" a college. Like the school Daminark and I go to, we're students in the college, which is a subset of the university. I don't know if other schools do this though.
 
I love the use of math terminology in normally-written English sentences.
 
For instance ?
 
Subset
 
5:59 PM
yo @Daminark can you recall the proof that compact operators on a hilbert space attain their norm
 
Eric said that Colleges are a subset of some Universities in the U.S.
 
I'm drawing a blank for some reason
 
Oh, I didn't even notice :p
 
Hah
 
Assuming separable, then take a sequence $w_n$ on the unit ball so that $\|Aw_n\| \to \|A\|$, and then use weak compactness to get that a subsequence $w_{n_j}$ converges weakly to some $w$.
 
6:02 PM
sure if separable
 
We hadn't done this but this is what I think is the case
 
i knew the separable case lol
 
But then $Aw_{n_j}\to Aw$, which should do it
Oh we never talked about non-separable spaces at all really
Hmm
 
yeah it was on charlie's final today
i did the separable case
but wasn't sure about in general
 
Huh, interesting
Oh wait a second
So we only proved Alaoglu for the separable case, but I think Schlag remarked that it holds in general by Tychonoff
So the closed unit ball is weak* sequentially compact always
But if your space is reflexive, this implies weak compactness
Or at least in a Hilbert space
So you should still be able to do the same argument
 
6:16 PM
oh duh
of course i completely forgot about banach alaoglu since I don't like tychonoff
 
Yeah, Schlag also didn't want to get into that stuff either, he was just like "Yeah this holds in general because of the dark arts"
 
i just always felt verrrry skeptical of it's statement
like i get the proof and worked through it and believe it from back when i took point set. But my brain keeps wanting to believe it's a load of bs
 
Hi chat
 
Yeah this does more to make me suspicious of the axiom of choice than anything else. Well ordering, Zorn's, bases of vector spaces, that's all fine, and the statement of the axiom of choice seems obvious
This is the one thing I've thus far seen which is making me kind of iffy
 
hello
yeah idk, axiom of choice was always fine with me, but some of it's equivalents seem like utter nonsense to me. But The same thing seems to be true of things like determinacy honestly
but honestly i dont care much about those kind of issues
 
6:22 PM
Lol that's like, fair
Last summer during the REU I was in an apartment with 4 friends and toward the later parts of the summer, our conversations were mostly centered around, what I guess you can call "the philosophy of set theory"
 
ew.
 
Canyon was kind of a constructivist so we often had those debates as well
It's admittedly of questionable productivity, like when actually working in math, it's entirely fair to just say that regardless of what you believe a set actually is, you can just say that this is a model of something and roll with it, letting truth apply merely within that model
But I've found such conversations to be fun nonetheless, if only to kill some time with
 
HI @Semi
 
7:16 PM
hi
does anyone want a cool math problem?
 
I don't mind hearing it
Not going to promise I'll try to solve it, though
 
Okay :P
Find, with proof, all integer solutions to the equation $4x^2 - y^2 - 15 = 0$, lying in the first quadrant
it's not that hard
 
Any progress on Sperner's Lemma ? @Meow
 
@Astyx My "theory" was as far as I got
sorry, I'm really demotivated right now :[
 
Want a hint ?
 
7:25 PM
$(2x+y)(2x-y) = 15 \leftrightarrow \{x,y>0\rightarrow 2x+y>2x-y \} x=2, y=1 \lor x=4, y=7$
 
@MeowMix Here's something real nice I solved a few days back. Prove that, for every $n$, there exists a circle (radius depending on $n$) on the plane which contains exactly $n$ lattice points - that is points of the form $(x, y)$ where $x$ and $y$ are integers.
Your proof should generalize to hyperspheres in $\Bbb R^k$.
 
Hey.
Someone mentioned a type of graph to me the other day, but I forgot what it was called. It looks like a plane divided into four quadrants, of which any one of those quadrants might be divided into four more quadrants, and so on.
 
@Lozansky yep
 
@MeowMix Prove that $7|(a^2+b^2)$ only when $7|a$ and $7|b$
 
user228700
Hello, everyone :-) I have a quick-ish question about the difference b/w a stationery point and a critical point.
 
7:40 PM
@BalarkaSen Ah, I think I've solved it. That's nice.
 
user228700
Is it that critical points also include those points at which the derivative is not defined (where as stationery points only include points at which the derivative is zero)
 
@Kaumudi No
 
user228700
Or is it that critical points are points where there is a potential maximum/minimum (the sign of derivative changes as the function passes through this point)? If so, I have a follow-up question; isn't this the same as an inflection point?
 
Not all critical points are potential maximum/minimums. There can be saddles, like $f(x) = x^3$
This is more apparent in higher dimensions
 
user228700
...what then, is the difference b/w stationery and critical points?
 
7:43 PM
@Akiva What did I expect... congrats :)
 
@Balarka Am I smart enough to know the solution? Or are you just pulling my tail.
 
@MeowMix It's not trivial. Think about it for a few days, which is what I did.
It is elementary however.
 
user228700
This is where I found the first way of differentiating the two terms:
 
user228700
2
A: What is the difference between stationary point and critical point in Calculus?

RedAll stationary points are critical points but not all critical points are stationary points. A more accurate definition of the two: Critical Point: Let $f$ be defined at $c.$ Then, we have critical point wherever $f '(c)= 0$ or wherever $f(c)$ is not differentiable (or equivalently, $f '(c)$ i...

 
I think it has to do with pythagorean theorem
I had an AMC question related to that...
 
user228700
7:46 PM
> Points where $f′(c)$ is not defined are called singular points and points where $f′(c)$ is 0 are called stationary points.
 
user228700
Is this correct?
 
well, pythagorean triples
 
@Kaumudi.H I have never heard of that definition of critical points. It's probably nonstandard.
 
user228700
@BalarkaSen Can u please tell me what the difference b/w critical and stationery points is, then?
 
@Kaumudi They're saying, critical points are points where the function has zero derivative. The trick is, this also includes points where the derivative does not exist!
Because if the derivative doesn't exist, vacuously the function has zero derivative.
 
7:48 PM
I think points on the boundary of the domain are also critical points?
 
This is an example of vacuous logic.
 
@BalarkaSen I don't think it works like that…
 
user228700
@BalarkaSen Right. So this is the difference b/w the two?
 
(I generally like to have an everywhere differentiable function whenever I talk about a critical point, but let's put that aside. All of this seems to be convention)
 
The definition of $f'=0$ is for all epsilon exists delta blah blah blah |stuff-0|<epsilon
which doesn't work for when the derivative doesn't exist
 
7:50 PM
Read the answer Kaumudi linked, @Akiva
I agree with that convention, but that's not what the answer's saying
They're saying, critical points are points where $f'$ makes sense and $f' = 0$ or points where $f'$ doesn't exist.
And they're saying stationary points are points where $f'$ makes sense and $f' = 0$
It's certainly totally nonstandard convention. I don't like it.
 
user228700
Uhhh, I'm just here trying to figure out what to do when asked to find critical points and stationery points.
 
They generally mean the same thing, but there's this dumb conventional difference I just explained to you.
 
user228700
Alright, then.
 
user228700
Thanks!
 

« first day (2417 days earlier)      last day (2605 days later) »