« first day (2850 days earlier)      last day (2174 days later) » 

7:00 AM
This is strange. I still have no idea how to visualize $\rm CP^2$. I used to imaging some sort of spherical symmetry but I guess that's gone.
 
@BalarkaSen shocking
 
The idea of $B^4$ with the boundary identified by a Hopf map makes sense, I guess
'cause the mirror image of the Hopf map is its negative
(in $\pi_3(S^2)$)
 
Yeah that's correct
It's the achirality of Hopf map at it's work
 
And $\pi_4(S^3)$ being $\Bbb Z_2$ means that $\Sigma H$ is not chiral somehow.
 
@Daminark Indeed
 
7:02 AM
I've got nothing to do for an hour (at which point I should go to sleep) so I should prob read up some AT actually
 
@BalarkaSen Achirality (or amphichirality if you like more lettters) is when it is its mirror image
 
At minimum I should get through the completely standard material on fundamental group/covering spaces/CW complexes
 
@AkivaWeinberger $\Bbb{CP}^2\#\Bbb{CP}^2$ and $\Bbb{CP}^2\#\overline{\Bbb{CP}^2}$ are nontrivially unlike. The latter is bounded by an orientable 5-manifold: $\Bbb{CP}^2\times I \setminus N_\epsilon(x_0 \times I)$
The former isn't.
Though it is bounded by a non-orientable 5-manifold by a souped-up construction
@AkivaWeinberger Oh I had forgotten that.
 
And $\rm\Bbb CP^2$ is the smallest thing not bounding anything, yeah?
 
Yeah.
Well, no, there's RP^2.
Smallest orientable thing not bounding anything
 
7:05 AM
@BalarkaSen Oh, that I didn't know
 
You can prove that
 
Yup. I leave it to you, Monsieur Akiva, as an exerciz
 
A sidequest!
(There is no main quest.)
(Life is the main quest.)
 
Accurate
Meanwhile, I'm upset that $\Bbb{RP}^3 \# \Bbb{RP}^3$ is not $S^1 \tilde{\times} S^2$.
That would have resembled $\Bbb{RP}^2\#\Bbb{RP}^2 \cong S^1 \tilde{\times} S^1$ (Klein bottle)
Such is life
 
7:09 AM
The latter fact is still really unintuitive to me
 
Oh I can tell you my intuition
It's what I was writing my proof for the wrong fact with
@AkivaWeinberger Consider the (open) Moebius strip. That's a line bundle $E$ over the circle which is a neighborhood of $\Bbb{RP}^1 \subset \Bbb{RP}^2$
 
Consider the quotient map $p : E \to E/E_0$ where $E_0 \subset E$ is the center circle of the Moebius strip, or the "zero section" of the line bundle. $E/E_0$ is nothing but $\Bbb R^2$, because the intervals in the Moebius strips are just untangling the lines through the origins in $\Bbb R^2$, which gets tangled after quotienting again
 
Oh so you glue two Möbii together?
 
You can do that to cook up a different proof
 
7:12 AM
So what does $E/E_0$ look like? A disk?
 
Yep.
 
Coo-well.
 
This map $p$ is called the blowdown operation. The reverse of this is called blowup. So $E$ is obtained from $\Bbb R^2$ by "blowing up" at the origin. Aka, $E \cong \Bbb R^2 \# \Bbb{RP}^2$ - because "untangling" the lines through the origins by distinguishing them by the direction vectors is precisely what RP^2 does.
Obtain the Klein bottle by simply one-point compactifying the fibers of $E$, which are open intervals. This has the effect that the first $\Bbb R^2$ factor of $\Bbb R^2 \# \Bbb{RP}^2$ gets compactified, but each line gets their own infinity. That's adding an $\Bbb{RP}^1$-at-infinity to $\Bbb R^2$, so we get the second factor replaced by $\Bbb{RP}^2 = \Bbb{R}^2 \cup \Bbb{RP}^1_\infty$.
So $K \cong \Bbb{RP}^2 \# \Bbb{RP}^2$.
That's my god-awful algebraic geometry proof of this fact. :D
That's what blowing up looks like, btw. Not hard to see that blowing up the plane at the origin gives the Moebius strip - just stack the lines through the origins one after another as it rotates.
Blowing up is the same thing as taking connected sum with a projective space.
 
@BalarkaSen I could never wrap my head around blow up except in the visual sense. All the formal stuff just swept by me.
 
I spent a shitload of time trying to prove the visual meaning agrees with the formal affine definition
 
7:24 AM
I do recall making Maple draw me a bunch of the curves before and after blow up, so I could get a better idea.
(also to get a idea of how it really did smooth out various singularities)
 
How many blowups you'd need to desingularize has always been a mystery to me
So in that sense I don't really get it
There are also singularities which never resolve after blowing up any number of times, I think. The Whitney umbrella
 
Okay this proof is clever, I like it
 
why are you doing analysis with my simplices
what is this vulgarity
 
Shit is about to get real when Balarka starts quoting authors in Russian
 
I mean this is more geometry than anything else, just done formally :P
One thing I don't like that this book does is this it doesn't use parentheses with fractions the way I'd want
So $(n/n+1) = \frac{n}{n+1}$
 
7:32 AM
V e r y f o r m a l
what the flip is this book you're readin
@AlessandroCodenotti In Soviet Russia, theorems prove you
 
This confused me earlier when it defined barycenter of the simplex $[p_0,\ldots,p_n]$ as $(\frac{1}{n+1}\sum_{i=0}^n p_i)$, and I thought it meant $\frac{1}{n} + 1$ and was so confused that it gave the barycenter of $[0,1]$ as $\frac{3}{2}$
This is Rotman
 
Hahahahah more like... Rot man
 
Rot in peace
 
gotteeem
 
Also wait who's the Russian author we're quoting?
 
7:35 AM
Alessandro probably refers to the page from Shafarevich in Russian that I posted above
 
Dostoevsky
 
Wait, when did I quote him
 
@BalarkaSen oh, I didn't know that's from Shafarevich
I was just joking, I wrote the first Russian author whose name's spelling I could remember
 
Oh lmaoo
 
Isn't there an English translation of Shafarevich by Miles Reid?
 
7:37 AM
There was this joke about a Russian going to an eye clinic and the doctor tells him to read some text on the board. He reads, and exclaims, "That's my brother's name!"
It was AFGTRSBCXZSKY
@AlessandroCodenotti Yep
 
Polish looks way more unpronunciable to me when written! But I met a lot of people from Poland when I was in Germany and it has an interesting sound
 
I was watching Beksinki's video diary a few days ago. It does have an interesting sound
 
I'm actually in Poland right now and the language is blowing my mind
(On the signage)
 
Can anyone give me the quick and dirty on what is flabby about a flabby sheaf?
 
I always forget that those are the same as flasque sheaves
 
7:47 AM
So what’s flasque about a flasque sheaf!
 
No idea honestly
 
That does actually seem like a weird name for them
I would have thought that being flabby should make the sheaf more rigid, but maybe my intuition is all wrong here
Maybe the idea is that most of the sheaf is "unnecessary" since whatever we care about, we should be able to figure out from smaller parts
so those larger parts just "flab about"
 
I like that answer
 
But I am not really sure how that is truly about the flasqueness
 
most of the names that seem to suggest some physical or visual interpretation make sense to me in some way, but I couldn’t picture what a flabby sheaf should look like
 
7:54 AM
Now try to visualize a perverse one :)
 
I’m pretty perverse most of the time it probably just looks like me
 
@BalarkaSen This is making sense, actually
Thanks
 
@Balarka seems this guy isn't a topologist, he explicitly stated and proved the gluing lemma early on and cites it often to show the maps he defines are continuous
 
What's the gluing llama?
^Speech-to-text
 
8:10 AM
@Daminark We cite it too. Implicitly.
 
If you write a space as a finite union of closed subspaces $X = \bigcup_{i=1}^n X_i$ and you have functions $f_i:X_i\to Y$ which agree on the overlap, then when you stitch the functions together you get a globally continuous function $f:X\to Y$
 
Hm
What if we say the subspaces are open instead?
 
Same thing holds
 
What if each $X_i$ is either open or closed
(…OK, I'll stop)
 
That's not gonna be true, take $X = A \cup A^c$ where $A$ is open and define unrelated continuous functions on each
 
8:14 AM
@Daminark I usually see it explicitly mentioned in the context of schemes instead of topological spaces
 
Ah, yeah
Oh, wait, if we have $X=\bigcup X_i$ with $X_i$ open, we don't even need it to be finite anymore, do we?
 
since there it is a lot less obvious
 
Yeah that's true
@Tobias huh, I'm not quite sure if I'm up to par on that business to understand what the statement is but in any event, what does it look like there exactly?
 
@AkivaWeinberger now I imagine a small llama spitting on the overlapping bits to glue the functions together
 
@Daminark I don't even recall. But similar, i.e. if you have a covering by open subschemes, then...
But once you make the notion of morphism stricter, it is clear that something might break.
 
8:18 AM
Gluing lemma is just the statement that the presheaf of continuous functions on a topological space satisfies the gluing axiom (so part of what makes it a sheaf)
The same thing can be said for the structure sheaf of a scheme.
 
 
3 hours later…
10:51 AM
The points of continuity of the function $f : \Bbb R \to \Bbb R$ defined by $f(x)=|x^2-1|$ if $x$ is irrational, and $0$ if $x$ is rational; are ....
I think they are $1$ and $-1$. Am I right?
Also, if we interchange the role of rational-irrational in definition of f, e.g, if $g(x)=|x^2-1|$ if $x$ is rational, and $0$ if $x$ is irrational, then also the point of continuity are $1,-1$ only, right?
 
@Silent Yeah
 
thank you!
 
11:15 AM
I can't see what is correct answer!
 
@Silent Did you notice that [B] and [C] are identical?
 
yes
 
Anyway, presumably you do know that Q implies P?
 
Yes. i thought about P does not imply Q, and thought that if there is a matrix such that $A^4=0$, then P true and Q false. So, P implies Q does not hold. But I can't digest that
 
Well, is it possible to have $A^4 = 0$ without $A^3 = 0$? Remember what size these matrices are
 
11:26 AM
@TobiasKildetoft Why can't that happen?
 
@Silent Think of characteristic polynomials
 
anyone aware of sum and product notation? (recursion - discrete maths)?
 
@DarkVampiricAbstractArtist sure
 
Thanks Tobias, i think this would be quick!
I need to write: 1/(6^4) + 1/(9^5) + 1/(12^6) + ... + 1/(33^13) as the product "Π"
 
Consider $(-1)^{\dfrac{(p-1)(p-2)}{2}}$. For $p = 2n$ we have
\begin{align}
(-1)^{\dfrac{(p-1)(p-2)}{2}} &= (-1)^{\dfrac{(2n-1)(2n-2)}{2}} \\
&= (-1)^{(2n-1)(n-1)} \\
&= (-1)^{2n^2 - n + 2n + 1} \\
&= i^{2(2n^2 + n + 1)} \\
&= i^{4n^2 + 2n + 2} \\
&= - i^{4n^2 + 2n} \\
&= - i^{2n} \\
&= - (i^2)^{n} \\
&= - (-1)^{n} \\
&= (-1)^{n+1} \\
&= (-1)(-1)^{n} \\
&= (-1)(i^2)^{n} \\
&= - i^p
\end{align}
and for $p = 2n + 1$ we have
\begin{align}
(-1)^{\dfrac{(p-1)(p-2)}{2}} &= (-1)^{\dfrac{(2n+1-1)(2n+1-2)}{2}} \\
 
11:35 AM
but i know that 10_Π_i=2 = 1/(3i)^i
 
@DarkVampiricAbstractArtist But that is not a product. I have no idea what sort of form that would take as a product
 
@TobiasKildetoft I'm talking about something like this
 
@DarkVampiricAbstractArtist But you wrote a sum, not a product
 
oh your right, its the sum, but i still dont understand
so, 10_Σ_i=2 = 1/(3i)^i
but however that comes out as 1/6^2+...+1/30^10
 
@DarkVampiricAbstractArtist That is because the exponents should be $2i$ rather than $i$
 
11:40 AM
since I need to write: 1/(6^4) + 1/(9^5) + 1/(12^6) + ... + 1/(33^13) for the sum "Σ"
 
woops, $i+2$ I mean
Also, you should go to 11, not to 10
 
oh yeah, my bad right there. so many silly/clumsy mistakes on my part.
11_Σ_i=4*
 
@TobiasKildetoft I tried for $A$ as well as $\det (tI-A^4)$. Since we are assuming $A^4=0$, we get $t^3+at^2+bt+c=t^3$. But then what?
 
@Silent I meant the characteristic polynomial for $A$ itself
 
ok
 
11:44 AM
@TobiasKildetoft i tried subbing it for i=5, but it came out 1/7^5
it should be 1/9^5
 
@DarkVampiricAbstractArtist subbing where?
You wanted 1/(3i)^(i+2)
 
Ah, I thought you meant 1/(i+2)^i
i was slightly confused there.
 
There, I had to change that twice because I kept writing it wrong
 
im a lot more worse, but thanks.
hmmm, when i sub in i=4, i get 1/12^4 ?
 
no
4 is not 4+2
 
11:48 AM
oh so i just sub in "2" to get "4"
 
as the exponent, yes
 
Thanks Tobias!
sorry to ask again. Does it looks like this? 13_Σ_i=4 1/(3i)^(i+2)
 
good morning everyone
 
No, you should sum from 2 to 11
 
11:56 AM
@TobiasKildetoft I can't see except eigenvalue should be $0$ since $A^4v=0v=\lambda^4v$ implies that $0$ is the only possible eigenvalue.
 
@Silent Ok, so what can the characteristic polynomial be?
 
@Silent as an example for operators, consider $D_x$ on the space of cubic polynomials. That's not (immediately) in matrix form but it is an example of a linear operator with $D_x^4 = 0$ but $D_x^3 \ne 0$
 
@TobiasKildetoft $t^3=0$ or $t^2=0$ or $t=0$.
 
@Silent No, precisely one of those can be the characteristic polynomial of a $3\times 3$ matrix
 
oh! $t^3=0$
 
11:58 AM
(also, the characteristic polynomial does not include a $=0$)
 
sorry!
So, how do we know that $A^4=0$ implies $A^3=0$?
 
@Silent We don't need to do so directly. We have just seen that any nilpotent matrix will have $t^3$ as its characteristic polynomial
So we use Cayley-Hamilton
 
ok
thank you very much!
 
but notice that the space I wrote is a 5 dimensional one so it's not a counterexample
 
@TobiasKildetoft, I came across this article which says 'This answer assumes that you are working over an algebraic closed field', and I found that 0 is only eigenvalue possible by the method described only!
I am working in rational numbers
 
12:08 PM
@Silent That comment is not really accurate. Changing to a larger field does not change the characteristic polynomial
 
ok, thanks again.
 
 
2 hours later…
2:12 PM
Hi, is is true that for every set all of its points are in the interior and/or in the boundary? I mean there can't be points in the set which are not at least in the interior and/or in the boundary, right?
 
Thx
 
2:29 PM
$A=A^\circ\sqcup\partial A$
 
I am looking at how to add more edges to a planar graph until it becomes maximal planar. Could someone point me in the right direction, e.g. give me a link to a relevant text? For simplicity, assume that I already have some planar embedding of the graph, which I can use to identify faces.
 
A set is the (disjoint) union of its interior and boundary
@Szabolcs A face is an $n$-gon for $n\ge3$, right?
If $n=3$ you can't do anything; if $n=4$ you can add one edge and then you can't do anything
 
@AkivaWeinberger Not necessarily in this case.
 
Is this true or not: Let $G$ be a group and $a,b\in G$. If $a^{17}=b^{17}$ and $a^{30}=b^{30}$ then $a=b$.
 
You're allowing multiple edges between the same pair of vertices?
 
2:32 PM
@AkivaWeinberger Here's one of the simplest cases when that doesn't work:
Suppose you started with a 4-cycle.
We added the edge 1-3
There's still the outer face left
 
user131753
@Silent In general, if $\gcd(m,n)=1$ and both $a^m=b^m$ and $a^n=b^n$ holds then $a=b$.
 
If you start with a 4-cycle, you have two faces
 
The outer face also consists of vertices 1,2,3,4. But we can't just add any edge now. We have to add 4-2 as 1-3 is already added.
 
@Szabolcs Oh, I see.
 
Another problem case is when we have a graph e.g. like this:
@Akiva Sorry, I should have said that I only consider simple graph (no multi-edges, no self-loops)
 
2:35 PM
@Szabolcs Right, I didn't think of that…
That's an interesting problem
 
Neither did I when I implemented what you originally said :-)
I am sure that there must be plenty written on this, but I can't really find the correct term to search for. "Triangulation" is also used for other things in the graph theory context.
 
@Silent Bézout says that $17x+30y=1$ for some integer $x$ and $y$.
$a=a^{17x+30y}=(a^{17})^x(a^{30})^y= (b^{17})^x(b^{30})^y=b^{17x+30y}=b$
@Szabolcs If I were in charge of naming things, I'd call it the "maximal planar supergraph"
(A superset is the opposite of a subset)
(Supergraph! *superman theme plays*)
 
@user170039, @AkivaWeinberger, thank you very much
@user170039, there are no points equidistant from $(-1,0)$, $(1,0)$ and $(0,1)$ in plane, right?
 
Let $g$ be defined on $\mathbb R$ by $g(1):=0$, and $g(x):=2$ if $x \neq 1$, and let $f(x):=x + 1$ for all $x\in \mathbb R$. Show that $\lim_{x\to 0}g \circ f \neq g( f(0))$. Why doesn’t this contradict Theorem 5.2.6?
5.2.6 Theorem Let $A, B \subset \mathbb R$ and let $f : A \to \mathbb R$ and $g : B \to \mathbb R$ be functions such that
$f (A) \subset B$. If $f$ is continuous at a point $c\in A$ and $g$ is continuous at $b=f(c)\in B$, then the
composition $g \circ f : A \to \mathbb R$ is continuous at $c$.
@silent
do you have any idea?
 
2:51 PM
If $\{E_n\}_{n=1}^\infty$ is a collection of sets and $\{F_n\}_{n=1}^\infty$ is the "corresponding" disjoint collection (i.e., $F_n = E_n - (\bigcup_{k=1}^{n-1} E_k)$, how is it possible that $\bigcup_{n=1}^\infty F_n = \bigcup_{n=1}^\infty E_n$ always holds? Certainly the set on the LHS is always contained in the RHS, but why does the other set inclusion hold?
 
@user170039, sorry to bother you, i think $(0,0)$ is such a point.
 
Can somebody give me a hint how to prove that a cuboid with empty interior is a null set? My definition is:

$N \subseteq \Bbb R^n$ is a null set, if for every $\varepsilon > 0$ there exists a sequence of *open* cuboids $(Q_\ell )_\ell$ in $\mathbb {R}^n$ such that $N \subseteq \bigcup _{\ell =1}^\infty Q_\ell$ and $\sum _{\ell = 1}^\infty \operatorname {vol}(Q_\ell ) < \varepsilon$
 
@user193319 I think if you have $x$ living in RHS, then you can take $n$ minimal such that $x\in E_n$. Then by definition $x\in F_n$.
 
Ah! Of course! I've been wracking my brain over this. Thanks!
 
@ManeeshNarayanan A function $h$ is continuous if and only if $\lim_{x \to x_0} h(x) = h(x_0)$. In your example $h=g \circ f$. But this is no contradiction since $h$ is not continuous at $x_0=0$.
 
3:03 PM
@ManeeshNarayanan Take $c=0$, then $f(c)=1$, but $g$ is not continuous at $1$. So, theorem 5.2.6's hypothesis is not satisfied.
 
can anyone recommend a good grad-level book on finite-dim Lin alg?
one that doesn't limit itself just to matrices
well I mean I know everything CAN be expressed in matrices in finite dim but
 
Hi. Given a simple graph, I'd like to construct such a random vector (corresponding to graph vertices) that the covariance between $X_i$ and $X_j$ would depend on $|\deg i - \deg j|$ but not any other graph characteristic (perhaps also size). I came up with mathb.in/25475?key=7f229bbb13600cbae10db4520034d70c7f9f335f, but the problem is that for typical degree distributions the covariances are "far" from zero. I'd like the covariance matrix to be sparse. Any ideas?
 
does Ted have videos on this
I meant doesn't limitself just to $\mathbb C^n$
 
@GFauxPas Well, i came across these: Lax's Linear Algebra, which he wrote for his next book functional analysis. Also, check evergreen book Finite Dimensional Vector Spaces by Halmos, which has companion problem-Sol book.
 
aaah YES Ted also stacks the arguments vertically in $f\begin{pmatrix} x \\ y \\ z \end{pmatrix}$
I thought I was weird for doing it
ah Ted's videos might be just what I'm looking for
thanks Silent
 
3:11 PM
@GFauxPas Oh! I think Roman's advanced linear algebra covers more than $\Bbb C^n$, even Dummit and Foote or Artin's abstract algebra texts cover more than $\Bbb C^n$
 
hm what are the kinds of things you want to learn when you say grad-level linear algebra?
 
I mean I guess you don't lose much generality because of isomorphism theorems
well let's see, I'm studying for my graduate written qualification exams, lets see what they say
Vector spaces, linear dependence, basis, dimension, inner product, linear transformation, systems of linear equations, matrices, determinants, ranks, eigenvalues, diagonalization of matrices, quadratics forms, symmetric and orthogonal transformations.
hmm that's mostly standard lin alg stuff
the site recommends Strang>
heard of it?
 
yes but ive never used it so i can't really comment on it :p i was just curious what you were referring to when you said grad-level linear algebra. personally a lot of stuff i think i kind of just picked up on the way - but it makes sense that you'd want a specific reference if you're studying for quals
 
yeah I have picked up a lot but I still need to study
I find functional analysis more interesting than finite-dim stuff but that's not what the exam is on :(
well the Halmos book is only $13 so might as well get it lol
I think Ted's videos look good :)
 
why do so many differential equations leave out important contexts?
it's like we're back to 200 years ago when every function was (assumed to be) analytic
 
3:23 PM
@LeakyNun for example? (i agree btw - although nowadays i barely rmb anything from diffeq)
 
see how many of them have even the domain of the function
 
oh
i think the problem is people who don't do math also need to know diffeq
and it's not taught properly
 
@loch that's like the whole problem with maths in high school
 
yes
i think the solution is to have people who actually know math to teach in highschool but im not sure if that is possible
 
@LeakyNun you can just assume everything is analytic using density arguments runs away
 
3:27 PM
but even in uni for other subjects -
for example i know someone doing biochemistry in my undergrad who had to play with things like dynamical systems for her module - and that's after two years of barely touching math - and they only had like a few classes on math for the semester because they also had to teach the biochemistry part of things
 
@GFauxPas under which metric?
 
it wus joke
 
(granted that course was optional -- so maybe a better example would be that they had to take a compulsory module where they mention things like fourier transforms, show them formulas but the whole thing just looks like black magic to them)
 
I mean there is the concept of weak solution
though I'm not sure how it works
 
3:45 PM
@LeakyNun, what is answer to this:
57 mins ago, by Silent
@user170039, there are no points equidistant from $(-1,0)$, $(1,0)$ and $(0,1)$ in plane, right?
 
the answer is $(0,0)$
 
I was thinking later that $(0,0)$ is the point
@LeakyNun But, how do we measure distance? $(0,0)$ is midpoint of $(-1,0)$ and $(1,0)$ so equidistant, but how is it have same distance wih $(0,1)$?
 
draw a picture
 
i drew,
 
the distnaces are all 1
 
3:49 PM
so are we using Eucledian metric?
 
of course
 
thank you!
 
Hi!
Anyone here familiar with graph theory?
Actually, you don't have to be familiar with graph theory to answer this basic question
Actually, forget about graph theory.
Suppose we have the following Voronoi diagram.
where the edges of the unbounded faces are connected to a vertex "at infinity"
Suppose that every vertex is of degree 3
If n is the number of vertices and e is the number of edges, why do we have this relation 3*n = 2*e?
You can think of this Voronoi diagram as a graph, of course!
How do you intuitively interpret the equation 3n = 2e, given my description of the graph?
Every vertex is incident to 3 edges (i.e. has degree 3) and every edge is incident to 2 vertices
 
4:05 PM
@BalarkaSen Hola senor.
 
I just would like to have a more intuitive interpretation of that equation
Why not e.g. 2*n = 3*e? Answering this question may be the key to intuitively understand the equation
 
@philmcole okay
@Silent okay
 
Note: this is a very basic math question! You should be able to answer, if you're a good mathematician, even without knowing anything about graph theory!
(Anyway, all these comments are, of course, mine.)
Suppose that, for each vertex, we had only one edge, and, for each edge, we had only vertex.
Maybe this is not a good example, given that an edge has always two incident vertices.
Let's suppose that a vertex has degree 1, i.e. it's incident to only one edge.
Now, given that a vertex can't be connected to more than one edge, then a vertex can't be incident to two (or more) different edges (a redudant conclusion). Hence, every edge has two different vertices from other edges. So, the total number of edges must be twice the number of vertices
 
@nbro if it is so easy, why are you asking us?
3
 
If n is the number of vertices and e is the number of edges, then it should be clear that, formally, we can represent the statement "the number of edges is twice the number of vertices" as 2*n = e.
 
4:28 PM
So, we can interpret 3*n = 2*e as the statement "twice the number of edges is thrice the number of vertices"
 
My guess would be to proceed by induction
Ie show that, if you have a Voronoi diagram with the requisite vertex-to-edge ratio, then adding another point gives a Voronoi diagram with the same ratio
But I frankly don’t care enough to pursue the argument
 
There are $7$ homomorphisms from $Z/14Z$ to $Z/7Z$, right?
 
@Semiclassical I don't want to prove it. I just would like to intuitively understand that equation. I think I would not be able to express that relation, if I was given the degree of each vertex, because I don't fully understand the equation
 
Induction is an explanation. It’s not always the most satisfying one, but it is an explanation
 
Clearly, every edge has always two vertices
The only thing that can vary is the degree of each vertex
There's a similar equation for a maximally triangulated graph, but between vertices and "faces"
 
4:49 PM
actually, maybe not yet, the user bar is red enough
 
@LeakyNun $\lim \limits_{x \to 1^-} \sum\limits_{n=0}^\infty (-1)^nx^{n²} = \frac{1}{2} \ $
My teacher has corrected this exercise today.
@LeakyNun Write $\sum_{n=0}^{+\infty}(-1)^nx^{n^2} = \sum_{n=0}^{+\infty}(x^{(2n)^2}-x^{(2n+1)^2})$
And then use the convexity $t \mapsto x^t$
to compare the series with \sum_n(x^{(2n)^2}-x^{(2n+2)^2})$
and \sum_n(x^{(2n-1)^2}-x^{(2n+1)^2})$
this series converge by alternating test
these*
 
Does anyone understand this question (not mine)?
0
Q: Why are stochastic processes governing the embedding of separable Hilbert spaces into a common non-separable Hilbert space?

zeraoulia rafikBy starting from a modeling platform in which multiple separable Hilbert spaces that are defined on top of the same vector space float over a infinite dimensional background separable Hilbert space that owns a companion non-separable Hilbert space, the individual separable Hilbert spaces are embe...

 
5:04 PM
O wow, a users from ages ago visited math chat. Must be an omen
 
*these series converge uniformly using property of alternating series so one can shift the limits
 
5:16 PM
Do you know how to prove for all function $f : I \to \mathbb R$, there exist $g,h$ satisfying the intermediate value theorem such that $f = g + h$ ?
 
Hi, I'm trying to find if there is a closed form for the number of rooted 2-dimensional polyominoes with n hexagonal cells, removing any rotational symmetries using any tile as the center of the rotation.
 
5:33 PM
$$\lim_{n \to \infty}n^2 (x^{\frac 1n}- x^{\frac 1{n+1}})$$
Its in $0.\infty$ form
What should be the strategy in $0.\infty$ type questions?
 
for $0 \times \infty$, we try to write it as $\dfrac{1}{1/0}\times \infty$ or $\infty \times \dfrac 1 \infty$
 
@Abcd try $\frac{x^{1/n}-x^{1/(n+1)}}{\frac 1{n^2}}$
 
abuse of notation but hopefully you know what i'm saying
 
Is this true: If $a_n>0$ for all $n$, then $\sum a_n$ converges implies $\sum a_n^2$ also converges.
 
@Silent en.wikipedia.org/wiki/Sequence_space#.E2.84.93p_spaces check under "$\ell_p$ spaces are increasing in $p$"
 
5:41 PM
@Silent and then?
 
wait no that's the wrong direction
 
Whom are you replying to @GFauxPas
me or Silent
 
Silent but the WP article has the wrong direction implication than what he was asking
@Abcd then L'Hopitals rule
 
@GFauxPas Without L Hospital?
 
um
 
5:44 PM
@GFauxPas other direction does not hold, since $\sum \frac1{n^2}$ converges while $\sum \frac1n$ does not.
 
@Silent right
lol, Hardy-Weinburg: $p^2 + 2pq + q^2 = 1$. Easy-Weinburg: $p = 1$
 
@GFauxPas lol :)
@GFauxPas Does this mean if $\sum a_n$ converges, $\sum a_n^2$ does, too?
for $a_n>0$?
 
For positive integers $m>n$, is there any way to express $(2^m-1)/(2^n-1)$ in terms of $m-n$ only?
 
no
 
@Silent: I can't tell if your question was answered or not. With the assumption $a_n\ge 0$, yes, it's true.
 
5:57 PM
Thank you !
 
Can you prove it?
 
Hey everybody!
 

« first day (2850 days earlier)      last day (2174 days later) »