« first day (2473 days earlier)      last day (2842 days later) » 

00:02
I love that book
on my reading list then
i wish i had enough time to read all the stuff i plan to read
@Lozansky Not really
Let's make this more general. Can we compare the eigenvalues of $A$ and $B:=A+kI$?
@Balarka people only live like a couple tens of thousands of days, if you read a book a day that's still not that many books compared to how many good books have been written :/
The eigenvalues $\lambda_A$ of $A$ are the solutions to $\det(A-\lambda_AI)=0$.
@Eric yep :(
00:11
The eigenvalues $\lambda_B$ of $B$ are the solutions of $\det(B-\lambda_BI)=\det(A+kI-\lambda_BI)={}$$\det(A+(k-\lambda_B)I)$.
I have a pretty restricted set of things I actually want to read or think I'll like reading but it still seems maddeningly huge
Thus, $\lambda_A=k-\lambda_B$.
Yeah same, my collection has become so big that it's really difficult to move around... which sucks as a college student who isnt living in places for extended periods of time... @Balarka
(Well, there's more than one eigenvalue of each matrix. But the equation above should give a one-to-one correspondence between them.)
@Lozansky This is true, $\det(A+B)\ne\det(A)+\det(B)$ in general.
@BalarkaSen Can you do a sanity check for me? From "Let's make this more general" - does what I wrote make sense?
Also a big problem for me is that I can't use bookmarks. I have to read a book more or less continuously, except for extreme cases where the shit's so psychedelic it doesn't matter if you start from page 50 and read back to page 30 or something
00:15
I have a bookmark collection but I don't use bookmarks lol
@AkivaWeinberger That sounds good to me.
Sanity checked :)
(by an insane person)
Yay! I always thought I might be sane
@BalarkaSen Haha
@AkivaWeinberger Shouldn't it be $\lambda_A=\lambda_B-k$?
@AkivaWeinberger You meant $\lambda_A = \lambda_B - k$, btw.
@Lozansky …Yes. I missed a minus sign.
@BalarkaSen I trusted youuu
00:20
Missed Lozanski's messeged
sorry to let ya down
i told you i was 50% insane
@AkivaWeinberger Thanks a lot, I'll try and finish the problem tomorrow. This should make it easier
G'night all
I need to sleep too. The sun's coming up.
lol
lunatic
"i'm deraaaanged"
nah too extreme
00:27
delusional
I got Lem's Futurological Congress saved. That's relatively short hence going to be the first thing I am going to read when I decide I want to read anything
01:06
I can't believe Balarka's un-sleep schedule.
$\log_{\sqrt[y]{z}}(z)=y.\quad\sqrt[\log_x(z)]{z}=x.$
These are trivial when read right.
(Found these on Math SE.)
I suppose substituting in $x=10$, $y=3$, $z=1000$ makes it all simpler.
$\log_{\sqrt[3]{1000}}(1000)=3.\quad\sqrt[\log_{10}(1000)]{1000}=10.$
 
1 hour later…
02:34
0
Q: Buffalo numbers and/or their variations

Eric StuckyThe Question (with details and generality removed) With proper capitalization, how many grammatical English sentences could "buffalo buffalo buffalo buffalo buffalo buffalo buffalo buffalo buffalo buffalo buffalo buffalo buffalo buffalo buffalo buffalo" possibly represent? Motivation The Eng...

yikes this took a long time to write up
02:51
@Mike I'm pretty sure theres K3's with infinite discrete automorphism group. No idea how one proves those things.
@PVAL-inactive I think K3s can never have holomorphic vector fields at least. I'm trying to come up with a plausible argument maybe using Bott residue but I can't figure it out yet.
03:59
can anyone apply for an NSF grant?
Hey everyone, quick question does anyone know a quick way to calculate $1823^{2903} \mod 2957$
Forever: you have to be citizen? Otherwise it's pretty grant-specific.
Apart from expanding $2903$ into $2^{11} + 2^9 + 2^8 + 2^6 +2^4 + 2^2 + 3$?
what if I want someone to pay me to do research, and not teach?
@Perturbative can you factor 2957?
Then you can use Euler's totient theorem or Fermat's little theorem
04:06
3 more hours until Netflix updates :)
<<<EXCITED
again, ??? it depends on the grant
@LeakyNun 2957 is prime though
oh...
I don't think there's other way then
04:48
Hey there everyone!
g'morning damin
Lol, 12 more minutes on my side before morning :P
But yeah, how's everything going?
haha same timezone :)
generally well
I came up with a counting question I like, and that's never a bad thing
Ah, nice! Care to share it? (I probably won't get it, I've got minimal combo experience thus far, but still)
it's up in the chat
not too far
(it's on the same screen for me, although I do have a fairly large screen)
morning semic :)
04:52
hi
Hey @Semi!
does the end of semester affect you at all? :P
I'm kinda annoyed at the commenter here: math.stackexchange.com/q/2277419/137524
@EricStucky Yes. Grading. Which I'll have to start again tomorrow.
o.O
did you just have late finals?
They did their final exam tonight.
04:53
yikes, okay
Oh god night exams..
I've got other grading I also need to finalize.
The compsci department here does that
6:30-9:30 pm, lol
Like in the intro to coding classes, exams are 7-9 the week before finals week
04:54
copper or bunny, semic?
3 hours? Whoa
I think they're both a bit exasperating
@Daminark math does it too sometimes
I'll go for the one who misses the point of the exercise.
@Eric ??? Wai...
04:55
Just because a function isn't integrable over the real line doesn't mean it doesn't have a Fourier transform.
it's infrequent, but they do it
Our finals (in the math dept) are 3h long as well, damin
So someone throwing up their hands and saying "I give up" when the objection they're raising isn't cogent is irritating.
and honestly most of my STEM classes at my undergrad were in that range too. It seems to just be a reasonable balance of time.
Ah, I guess that was not really my irritation with that comment, but yeah :/
Huh, here all finals in their scheduled time are 2 hours
Of course there's some variation, Eric's professor last year gave them unlimited time (unfortunately we won't get that)
04:58
oh yeah i stayed for like 5
someone stayed literally all day
I know
I talked to him right afterwards and was like ???
(If one is clever, one only actually needs the Dirac comb to do the problem: Any periodic extension of a function supported on a interval can be written as the convolution of a Dirac comb and the function itself. So Fourier transforming a periodic extension will just be a weighted Dirac comb.)
But yeah thing is, because the school decided for some reason they wanted us to go away on Friday instead of Saturday, the finals were fucked, and among other things, we're having ours late.
He spent a bunch of time contour integrating a thing @Daminark, but to integrate it you could just use some basic fourier stuff and calculate it in a couple seconds
Yeah we were both doing the thing with our physics TA and complex analysis
He actually knew what was happening :P
05:00
lol
But yeah, Marianna did mention that while that time didn't go to waste because of this, asking a TA to stay for 8 hours is too much, so we won't get unlimited
maybe she just didn't feel bad asking that of alan cause she's his adviser lol
Lol, I wouldn't've expected that. Like, Soug didn't want to make our TA grade the entirety of our psets. To be fair, asking that would be cruel in general, but even though he's her adviser, still only said to grade a tenth of it
Also, fwiw, Marianna probably thought that the majority of the class should've finished in 2 hours, maybe a few people staying for 3 or 4 :P
Hey @Alessandro!
I stayed a little long because I didn't study the Marcinkiewicz interpolation theorem and so had to figure out how to prove it
that's what i spent like 80% of my time on the exam doing
05:19
ooh boy
I need to talk to Ted, but he seems gone
\:
How's it going @Alessandro?
And aw @EricS
Pretty well
What about you?
Sorry for coming in guns hot with a question, but I dont feel like its worthy of a post on MSE - are minimal primes preserved through a homeomorphism of spectrum?
05:28
this is probably a no brainer but I have 0 topology background
the only chat regulars who I know to be equipped to answer that are MikeM and Balarka (maybe Danu?), but none of them are on :/
No problem :) thank you for that information!
we are talking in context of optimization stuff
I guess I'll just do what I should do... and try to prove it : 0
05:32
can somevody explain me what is a mode of the function (couldn't dind any article on google)
an somevody explain me what is a mode of the function (couldn't dind any article on google)
is the function a p.d.f. for a random variable?
Well here is the full defination
Actually I am struggling hard to understand this paper:
Can soembody ehlp?
it's a short paper but very heavy, can somebody explain part2 and 2.2
06:34
hello
can you help me on metric space ?
probably not
but you can try
How we prove that $d(A,B)=\inf_{x\in B} d(\{x\},A)$ ?
06:52
@Vrouvrou what is the definition of $d(A,B)$?
$d(A,B)=\inf_{a\in A, b\in B} d(a,b)$ that's what i know
@Vrouvrou and what is the definition of $d(x,A)$?
0
Q: A Simple Question about Poles

EsterIf $f$ be a meromorphic function on the complex plane and $(1-|z|)|f(z)|$ $\longrightarrow$ $0$ as $|z|$ $\longrightarrow$ 1 then show that $f$ does not have any pole on the unit circle. This looks very simple but I somehow have not able to prove it rigorously. Thanks for any help.

@LeakyNun $\inf_{a\in A} d(x,a)$
@Vrouvrou so have I solved your problem?
06:57
just this ?
try applying the definitions to your equation
Hello everyone! (Briefly, though, I gotta go to bed soon)
@Vrouvrou tu me comprends?
Hello, Demonark
oui, j'ai compris, je pensais que c'était plus compliqué
alors j'ai resolve ton probleme?
07:01
How's it going?
je pense oui, on applique juste les définitions
@Vrouvrou (y)
@LeakyNun do you know how we prove that $$A ~\text{bounded}~ \Longleftrightarrow \exists a\in E, \exists r\in]0,+\infty[; A\subset B(a,r)$$
what is $E$?
what is the definition of bounded?
$(E,d)$ is a metric space
07:13
sorry I'll be back later
answer my question in the meantime
@LeakyNun we say that a set is bouded if it's diameter is finite
how do you define the diameter?
wow, chat's getting crowded O.O
how y'all doin this morning?
hey guys
does S_3 isomorphic to the group of rotation of angle 120 degrees and reflection?
07:28
Do you think yes or no?
okay, so you need an isomorphism
S_3 has two generators with orders 2 and 3 and they do not commute
so I guessed the isomorphism
but it's just really a check
so I asked
whether they are isomorphic or not
07:31
it's pretty much just a check
that your 'isomorphism' is, actually a bijective homomorphism
what is your isomorphism?
@LeakyNun $diam(E)=\sup d(x,y)$
@EricStucky S_3=<T,R> and the order of T is 2 and the order of R is 3
so send T->reflection
and R->rotation 120
okay, well, does it work? I mean, you have $S_3=\{I, T, T^2, R, RT, RT^2\}$, so define $f(T^2)=f(T)^2$ and $f(RT)=f(R)f(T)$ and $f(RT^2)$ to $f(R)f(T^2)$. Then is it true that $f(xy)=f(x)f(y)$ for any $x$ and $y$?
That's what you have to check.
The group is small enough that you can just do all the cases of $x$ and $y$; although some of them you can obviously skip, like when $x=I$.
@AkivaWeinberger I think I can draw mine from memory (maybe not the little things in the corner)
07:47
@Vrouvrou let $a$ be an arbitrary point in $A$. Verify that $B(a,diam(A))$ satisfies your condition.
Hi chat
@Vrouvrou Didn't I aready help you with this question a month ago or so ?
@Astyx ???? no
Oh it's because it was the same idea of the last question of yours I answered
07:50
Hi @MikeMiller.
@Astyx how is that the same idea lol
@Vrouvrou tu me comprends?
non pas encor , je vais ecrire pour voir si j'ai compris
Wait I might be confusing you for someone else
Oh no it was you, this question
@MikeMiller Is it hard to see whether the cohomology of $SO(4)/T^2$ has any torsion besides 2-torsion (which I think it should have for... reasons)?
@BalarkaSen Do you know spectral sequences well enough to say something about this ^?
@Astyx it is not the same question it has no relation
07:59
You mean torsion in... cohomology of SO(4)/T^2?
Yes (added it in the original message)
I was talking about this question, I hadn't notciced it had been answered since @Vrouvrou
@Danu How's T^2 acting on SO(4) here? As a subgroup (which subgroup?)?
@LeakyNun i must go, thank you for your help
@Vrouvrou alright
08:02
@BalarkaSen A maximal torus
@Danu Hmm. So in your bundle, T^2 --> SO(4) --> SO(4)/T^2, is the monodromy action trivial? That is, does $\pi_1$(SO(4)/T^2) act trivially on $H_*$(T^2)?
I don't know :\
I only just discovered that this could help me out so I haven't thought about this at all
That's essential for me to do spectral sequences :P
(I don't know either)
@Ted Which video of yours should I start with for manifolds ? (there are so many)
08:23
@Astyx Let me look.
You already know a decent amount of multivariable calc, right?
The two implicit function theorem lectures are great, IMO
(The 2nd part introduces manifolds formally)
 
1 hour later…
09:47
Hey @Bala
@Danu as well
Chat in general, I suppose
Hey @Krijn
0
Q: Construct a sequence using the fourier transform

user8469759Consider the Fourier transform defined as $$ \mathcal{F}[f](p) = \int_{-\infty}^{+\infty}f(x)e^{-j2\pi xp}dx $$ Assume that $$ f(x) = \left\{ \begin{array}{ll} 1 & |x| < \frac{1}{2} \\ 0 & \text{otherwise} \end{array} \right. $$ under such hypothesis we now that $$ \mathcal{F}[f](p) = \hat{...

If you want to have a look
10:03
Blegh, googling for hours now to find the right term
for what
3
Q: Complex differential geometry, complex algebraic geometry, and complex analytic geometry

Jason BourneFrom my limited understanding, complex differential geometry studies the differential geometry of complex manifolds, and complex algebraic geometry studies algebraic geometry where the underlying field is the field of complex numbers. I have come across the term complex analytic geometry in a num...

My question was not downvoted, but it was closed. I have gotten the answer I need, but I don't think it should be closed. It's not too broad. It's a very specific question, like asking what the colour of your toothbrush is.
@BalarkaSen My supervisor told me $SO(4)/T^2$ should be $S^2\times S^2$ (somethingsomething universal cover to go to $SU(2)\times SU(2)$ something something)
I don't really understand it
@Danu Huh. Shrug.
(so no torsion at all)
10:18
@BalarkaSen Talking about shrugging, I can make the bones in my neck make a sound by moving my head.
10:51
@Danu : See the discription of SO(4) here: math.stackexchange.com/questions/46131/… . In particular, SO(4) is isomorphic to $SU(2) \times SU(2)/ \mathbb Z_2$ and $SO(4)/T^2$ is really like $SU(2)/ U(1) \times SU(2) \times U(1)$, and $SU(2) /U(1) \cong \mathbb S^2$ is really the Hopf fibration.
can I act S_3 on IR^2?
@mathvc_ S3 is the group of symmetries of an equilateral triangle. So it's a subgroup of Isom(R^2), right? So you have an action by isometries on R^2.
is it rotation plus reflection?
Rotation and reflections are both isometries of R^2.
I don't understand the question.
@BalarkaSen true i mean can we act S_3 on IR^2 by rotation and reflection?
10:59
Yes, that's exactly the consequence of it being the symmetry group of the triangle, like I said.
thanks yup
I saw your question earlier about constructing a space with S3 fundamental group. Are you trying to quotient by this? That wouldn't do, because this is not a free action.
@JohnMa I guess you want $SU(2)/U(1)$ in that second factor too :) I figured that this is what you need to do but I am not sure how to prove it :P
Hello everyone
How to do this^?
Any hints?
f(x+y) = f(x)f(y) is satisfied if f(x) = e^{ax}
sorry forget about it
not sure that's useful
11:06
@user8469759 but $f'(0)=11$
that would mean that a = 11
@Fawad f(x+3) = f(x)f(3) = 3f(x). Differentiate both sides to get f'(x+3) = 3f'(x). Substitute x=0.
@LeakyNun do I need to check is it odd function or even function?
@Fawad why do you need to?
f'(-3)=11/3 but we need f'(3)
11:12
@Fawad read my instructions again.
Ok. 👍 nice
3
Q: What does happen if we don't define $C^\infty(p)$ as the set of germs of the real valued functions?

user220178I've seen in some books and notes that in order to define the tangent vector $v:C^\infty(p)\to\mathbb R$, the author defines $C^\infty(p)$ as the set of real valued functions $f:M\to\mathbb R$ such that does there exist an open set $U\subseteq M$ containing $p$ and $f|_U$ is smooth and then defin...

@Fawad a harder follow-up question: can you prove that f must be in the form of e^(ax)?
@user8469759 ^
@Danu I did not try all the detail yet, but you can (1) consider the $(p,p) : SU(2) \times SU(2) \to \mathbb{P}^1 \times \mathbb{P}^1$, where $p$ is the usual Hopf fibration, and (2) show that the map descends to $SO(4)$, and then (3) check the maximal torus is the stabilizer.
:37347487 I'm asking you, not math.SE
11:21
@Astyx The corner is just an equilateral triangle with the middle missing (Sierpinski gasket/triforce)
@JohnMa I'll give it a try in a bit
I'm kind of stuck on this question.. Can anyone solve it? How many seven digits no. can be made from 1,1,2,2,3,3,3,4,4?
My solution is 9x8x7x6x5x4x3/2/2/3/2=3780 but i don't kniow if its is correct
11:38
Guys, say we have
$$
x(t)=Ae^{i\omega t}+Be^{-i\omega t},
$$
where $x(t)$ has to be real ($\omega$ and $t$ are real also). I’m guessing we could say then that $A^*=B$. I would like to write this in the form
$$
x(t)=A\cos(\omega t+\phi).
$$
I started as follows:
$$
x(t)=(A+B)\cos\omega t+i(A-B)\sin\omega t.
$$
So we have: $A+B=E\cos\phi$ and $i(A-B)=-E\sin\phi\iff A-B=iE\sin\phi$. So it follows that
$$
\tan\phi=\frac{-i(A-B)}{A+B}=\frac{-iA+iB}{A+B},
$$
and it's easy to show that $\tan\phi$ is indeed real:
never mind I'll ask somewhere else!
If the probability of scoring a hit is 0.3, army A has 2 units, and army B has 1 unit, then A has a 0.86839 chance of winning, B has a 0.09213 chance of winning and there is a 0.03948 chance of a draw.

Would you guys please help me to understand answer of win,loss and draw probability values?
11:53
@LeakyNun sorry(mom called for lunch),don't know how to prove ..trying..
@LeakyNun I got $f'(0)=f(0)=1$
Can I replace 0 with x ?
you can't
How to proceed then?
find f(2) in terms of f(1)
Oh got. $f(2)=f^2(1)$ similarly $f(n)=f^n(1)$ then?
I see this pattern^
12:03
$f'(n)=nf(1)f'(1)$
Since it is defferentiable and if we can prove f'(x)=f(x) then ..
why can you prove f'(x) = f(x)?
Never mind. How to proceed then?
find f(1/2) in terms of f(1)
12:08
@LeakyNun $f^2\left(\dfrac 12\right)=f(1)$
then?
Ooh. That similarly thing is wrong
@LeakyNun then what?
you haven't expressed f(1/2) in terms of f(1)
$f(1/2)=\sqrt {f(1)}$
now f(1/n)
12:13
Maybe it's the way I set my fonts in my browser, but Leaky Nun's avatar appears extremely long compared to others.
It sort of makes Leaky Nun appear as two avatars instead of one, and makes the whole room look very ugly.
$f(\frac{1}{n})=f^{\frac{1}{n}} (1)$
@LeakyNun
then f(p/q) where p and q are integers
@LeakyNun replace 1/n with p/q ?
@LeakyNun in terms of?
12:19
f(1)
@LeakyNun don't know how to do but I think it Is f(p/q)=f^{p/q} (1) by observing pattern
you need to prove it
Don't know how to prove.
can you prove f(nx) = [f(x)]^n where n is an integer?
> Writing $\begin{bmatrix}x\\y\end{bmatrix}\in\Bbb R^3\times\Bbb R^3$, show that the equations$$\|x\|^2=\|y\|^2=1 \quad\text{and}\quad x\cdot y=0$$define a $3$-dimensional manifold in $\Bbb R^6$. Give a geometric interpretation of this manifold.
@BalarkaSen Did you get that this was $\Bbb R\rm P^3$?
(It's problem 6.3.12 from Ted)
12:31
@AkivaWeinberger I haven't thought about it, but $\|x\|^2 = \|y\|^2 = 1$ in $(\Bbb R^3)^2$ is $S^2 \times S^2$, right?
So I have to figure out what $x \cdot y = 0$ means
@LeakyNun let f(a+b)=f(a)f(b) then f(2a)=f^2 (a) , f(3a)=f^3(a)....f(na)=f^n(a)
yes
now put x=1/q
@BalarkaSen Yeah.
Hi there
I feel very stupid about this question I'm about to ask
$f(p/q)=f^p(1/q)$
12:34
Turns out I don't even need the cohomology of $SO(4)/T^2$. Just that of $SO(4)$ :D
and can you find f(1/q)?
When evaluating $\int(9\sin(5x+\pi))dx$, why does WolframAlpha tell me to convert it to $\int(-9\sin(5x))dx$ and determine this integral instead?
@LeakyNun $f^\frac 1q (1)$
And so $f(p/q)={}\dotsb$
@Steve Because $\sin(A+\pi)$ is the same as $-\sin(A)$
so it probably is programmed to do that substitution automatically.
Oh ok, I must have forgotten that. Why is $\sin(A+\pi)=-\sin(A)$?

@AkivaWeinberger
12:39
Hmm $f(p/q)=f^{p/q} (1) $
@Steve Think about the unit circle. Alternatively, use the sum formula for sines.
Sin is negative in third quadrant
@LeakyNun We're assuming $f$ is continuous, right?
@LeakyNun then? You take much time bro
12:41
I thought Akiva was going to continue
@Fawad So, for any rational, $f(r )=f^r(1)$, right?
@AkivaWeinberger Thinking of $S^2 \times S^2$ as a bundle over the x-copy of $S^2$, cutting out by $x \cdot y = 0$ probably means it's a $S^2$(y-sphere) $\cap T_x S^2$-bundle over the x-sphere $S^2$. So I suppose it's the unit tangent bundle of $S^2$, which is $\Bbb{RP}^3$.
Right
I hope that's a right intuition.
@BalarkaSen That's probably right, though I don't know anything about bundles. I was just thinking that it's $SO(3)$ (it's not hard to biject our manifold to the set of rotations), and $SO(3)=\Bbb R\rm P^3$.
@Fawad What about $f(x)$ for irrational $x$? Remember that $f$ is continuous
12:43
Yeah I suspect that's a more definite way to do it, but I am bad at matrices :P
Firsf make a guess and then prove it @Fawad
@BalarkaSen No matrices required. Think geometrically.
Right, of course. Thanks
Each element is an ordered pair of orthogonal points on the sphere, right?
Call $((1,0,0),(0,1,0))$ the "standard" pair of orthogonal points on the sphere
12:44
Yeah, got it.
Yeah, you rotate it to the element
By picking a normal to Span(x, y), you just have the space of orthonormal bases on R^3, which is SO(3) like you said
@AkivaWeinberger same for irrational.
Good proof.
@BalarkaSen Right, same idea
@Fawad How would you prove it?
Hint: $f(3)=f^3(1)$, $~f(3.1)=f^{3.1}(1)$, $~f(3.14)=f^{3.14}(1)$, etc. (due to what we proved for the rationals).
And $f$ is continuous.
12:47
@AkivaWeinberger by observing pattern?
Hello. I am reading this proof on the center of the dihedral group: ysharifi.wordpress.com/2011/02/02/center-of-dihedral-groups One part I am having trouble understanding is why $o(b) = n$ (the order of the rotation $b$) and $n|2j$ entail that either $j=0$ or $2j=n$.
@Fawad (@LeakyNun) I need to go, but you should be able to figure out how to go from here
@Fawad do you know about the result that says every irrational number is the limit of a sequence of rational numbers?
@LeakyNun yes. As dogatemy slowly wanted to show $f(\pi)=f^\pi (1)$
yes
so what can you conclude?
12:58
For irrational numbers also $f(\text{irrational})=f^{\text{irrational}}(1)$
can you write it more rigorously?
$f(j)=f^j(1)$ for all real $j$

« first day (2473 days earlier)      last day (2842 days later) »