« first day (4391 days earlier)      last day (631 days later) » 

12:05 AM
The view out my window right now:
 
A little drama.
 
 
3 hours later…
2:39 AM
@XanderHenderson that's a rainbow.
newbie
j/k
 
3:15 AM
hint: don't rub your eyes after chopping jalapeños
 
hey guys
is there a way to find all numbers (between 1 and 100 inclusive) which give a remainder of 9 when divided by 11?
 
3:28 AM
@TedShifrin Yup I agree, Ted. It's number 6 in Section 1.1 of Bartle's Introduction to Real Analysis. Part (a) and (b) of the same problem are to "show" the symmetric difference is given by those formulas.
 
$9+11n$?
 
@Koro why would it not be valid?
 
4:05 AM
Of course
 
the communications have become very cryptic
 
My of course is way out of place.
 
No thanks
 
A.G
Good morning mr shifrin..I am a bsc first year student. I was reading chp 5 in your book on multivariable calc and didn’t understand the intuition behind representing a quadratic form as a product of matrices..cld you pls give me some insight?
 
You mean $Ax\cdot x$ for a (symmetric) matrix $A$?
 
A.G
4:17 AM
Yes
 
It’s a quadratic in $x$ …. So you need two $x$s with coefficients.
 
A.G
I get that..I get logically why you can do that..but how does it help, what do we do it for?
I get the procedure for doing it
 
Because you can do the algebra of completing the square by matrix row reduction.
 
A.G
Ohh..and why do we require A to be symmetric?
 
You also can use the linear algebra of the matrix $A$ (either $LDL^\top$ or eigenvalues) to understand the quadratic form and hence classify critical points. Symmetry gets you that and also it makes the matrix unique.
 
4:30 AM
what was the of course referring to?
 
Some of us do, but that's hardly true of everybody
 
-_-
i may have to take my first algebraic geometry course when my next semester starts, the instructor for the course follows hartshorne , the course is called 'introduction to AG'
(aka in september)
its between that and a PDE course on hyperbolic PDEs and lorentzian geometry
im pretty much a noob at comm alg though so id need to go through atiyah mcdonald and likely some other stuff if I want to take this AG course
 
@porridgemathematics to finding the numbers $9\pmod{11}$.
To do Hartshorne you need your CA to be super solid. Maybe your lecturer will be gentler, but the book is not.
I took Hartshorne's course having had both CA and courses on Riemann surfaces and complex manifolds, so I had some idea of (smooth) algebraic varieties when I started.
 
4:51 AM
@porridgemathematics because someone objected to me that at some point F_{i+1}\F_i may be empty.
And I don’t think the objection makes sense here considering the way I chose F_i’ s.
8
Q: Prove that an infinite sigma algebra contains an infinite sequence of disjoint sets and is uncountable

KeithThe following question is from Folland Real Analysis, chapter 1 problem 3. Let $\mathcal{M}$ be an infinite $\sigma$-algebra. Prove that a. $\mathcal{M}$ contains an infinite sequence of disjoint sets. b. $\text{card}(\mathcal{M}) \ge \mathfrak{c}$. This is the problem I'm totally stuck at. Fi...

 
that wasn't what the commissioner saw!
 
The answer here uses a very complicated approach to the construction.
 
For the following sentences in Axler's linear algebra done right
 
do anyone think that my construction of the disjoint sets is not correct?
 
He states that if we treat p as an element of P(C), then complex roots will come in pairs
 
4:59 AM
@Koro Let $C_x = \{A \in \Sigma | x \in A \}$ and take the intersection.
 
Or at least that is my interpretation of his sentence
however here I have found that it doesn't necessary hold true for polynomials with complex coefficients
is someone able to explain what axler is trying to say here
I can attach 4.14 and 4.15 if needed
 
for a polynomial over the reals, if $\lambda $ is a root then so is $\overline{\lambda}$.
 
he's talking about polynomials with real coefficients, whose non-real roots do come in pairs as he describes
 
totally unreal
 
he's not saying anything about polynomials whose coefficients might not be real, where you are right, the roots can be anything
 
5:02 AM
zeroing in on the solution
 
Ty, I guess the line "Thus if the factorization of p as an element of P(C) includes terms of the form (x-λ) with λ a nonreal complex number, then (x-$\overline{\lambda}$) is also a term in the factorization" threw me off
 
oh yeah, i see. "as an element of P(C)" there is referring to the factorization in the world of P(C), but not changing that the p being factored has real coefficients.
i agree that it could have been written better. it pains me to say that about axler.
 
Ohhh
Ok, thank you
 
6:02 AM
For the following proof
How does Axler get the equation 0 = Im q(x) = (Im a0) + ... + (Im a n-2)x^n-2
I'm not the most sure how Im works with a polynomial but I'm assuming it is just applied to the coefficients (please correct me if I'm wrong)
but I'm not sure where he gets the 0 from on the left side
 
@TedShifrin yeah that makes sense, i've had courses on the riemann surfaces and complex manifolds, but not any on CA, there is a graduate course on CA being offered but my supervisor seems keen on me taking the AG course.. I'll have a chat with him about my concerns
i could take both concurrently, but itd be a little awkward, might be better if I took CA this semester and the AG course next semester
@Koro ah yeah, it doesnt quite work because $E_2 $ not equal to $E_1$ doesn't mean it cant be a subset of $E_1$, I didn't read carefully enough.
 
6:17 AM
i game on a mac: this is related to the last question somewhat. he's applying Im to the complex number p(x) and using properties of Im. Im is additive, so Im (polynomial) is the sum of Im(each term). Im is also real-linear, so if x is real then Im(a x^k) = Im(a) x^k for nonnegative integers k and complex a
if x were not assumed to be real, there would be no reason to expect Im (sum a_k x^k) to be the sum over k of Im(a_k) x^k. in general you could say that Im(sum a_k x^k) was sum Im(a_k x^k) but would not be able to pull the x^k out of Im in general
 
6:32 AM
@Koro as @copper.hat pointed out, the standard way to see that you cant have countably infinite sigma algebras is by supposing that you had some countably infinite sigma algebra, and then taking advantage of this by defining $[x]$ to be the intersection of all measurable sets containing $x$, which is necessarily in the sigma algebra (since its a countable intersection), and since each $[x]$ is the smallest measurable set containing $x$ these sets partition the sigma algebra
but if you only had finitely many of these sets, your sigma algebra would be finite, so you have some partition of your sigma algebra that is infinite, which means your sigma algebra has cardinality $\geq 2^{\aleph_0}$
 
 
6 hours later…
12:04 PM
Does anyone know why $z^1=\dots=z^m=0$? Here $NM$ is the normal bundle.
 
12:17 PM
that condition on $\Phi(x,v)$ is equivalent to $v$ being orthogonal to $E_1\vert_x,\dotsc,E_m\vert_x$ and these form a basis of $T_xM$ by construction
 
hm, I didn't realise that $E_1,\dots, E_m$ was a frame for $TM$, let me see
oh right
I forgot how the subspace of an embedded manifold looked like
thanks for clarifying
 
12:59 PM
Yeah, perhaps it is simpler to formulate this without transporting everything back via charts. If $M$ is an $m$-dimensional submanifold of $\mathbb{R}^n$, then $M$ in $\mathbb{R}^n$ locally looks like $\mathbb{R}^m\times\{0\}^{n-m}$ in $\mathbb{R}^n$.
The tangent bundles then look like $(\mathbb{R}^m\times\{0\}^{n-m})\times(\mathbb{R}^m\times\{0\}^{n-m})$ in $\mathbb{R}^n\times\mathbb{R}^n$ (with the usual canonical identifications), so the corresponding normal bundle is $(\mathbb{R}^m\times\{0\}^{n-m})\times(\{0\}^m\times\mathbb{R}^{n-m})$ in $\mathbb{R}^n\times\mathbb{R}^n$, which is an $
 
 
1 hour later…
2:01 PM
are varieties noetherian topological spaces
 
2:19 PM
@Thorgott Yea that is a lot more transparent and I will remember it that way, thanks!
 
yesterday, by 299792458
Hello respected mathematicians. I am here with a text-reference request, for singular value decomposition applied to curve fitting problems. I can find plenty of online resources and lecture notes for this, but can this be traced to some comprehensive treatment in some linear algebra textbook. Gilbert Strang has the rudiments put together in the appendix, but that hardly qualifies as a detailed treatment.
Respected all, any pointers would be very useful.
yesterday, by 299792458
Any other book/source that you may be aware of, for this?
 
2:48 PM
@monoidaltransform what kinds of varieties
 
@monoidaltransform yes. a variety is (among other things) a finite type scheme over a field, such schemes are noetherian and noetherian schemes are noetherian as top spaces. If you mean classical varieties, then the answer is also yes
although of course a variety over C with the analytic topology is not Noetherian. But it is Noetherian in the Zariski topology
 
3:30 PM
@leslietownes Thank you, may I also ask where he got the 0 on the left hand side from? Is it the fact that q(x) is real for all real inputs and thus Im q(x)= 0 as well?
 
Yes.
 
4:20 PM
Hi!
> Let $\omega=-\frac 12+i\frac {√3}{2}$ and $S$ denote the set of all the complex numbers in the argand plane of the form $a+b\omega+c\omega^2$, where $a,b$ and $c$ belong to $[0,1]$. Then find the area and perimeter traced by $S$.
How do I approach this? On simplifying, it becomes $\frac{2a-b-c}{2}+i\frac{\sqrt3(b-c)}{2}$
 
4:57 PM
Think about two linearly independent vectors $v,w$ in the plane. What does $\{sv+tw: 0\le s,t\le 1\}$ look like?
 
5:12 PM
@TedShifrin $sv+tw$ will represent some vectors in the plane of vectors $v,w$ for different values of $s,t$? however i don't understand how would they look like
 
You need to figure out that picture. There’s a very simple geometric explanation.
Draw pictures.
 
 
1 hour later…
6:24 PM
@TedShifrin i tried drawing, that's like a lot of vectors lying b/w vectors $v$ and $w$, i tried by taking example of $v=i$ and $w=j$ and different values of $s$ and $t$, can you please help further?
 
Not between. Think about whst happens when $a,b=0,1$.
 
also, i am in high school (in case it requires some more knowledge of vectors) or probably i am dumb :/
 
No, no more knowledge. Just persistence.
 
@TedShifrin do you mean $s,t$?
 
Oh sorry. Yes.
 
6:28 PM
if both $s,t=0$ then it is a null vector; if both 1 then sum of $v$ and $w$ and if either of them 1 and the other 0, then either $v$ or $w$ depending on which one is 0 and 1
 
So you should see it now. You have $0$, $v$, $w$, and $v+w$.
 
triangle?
 
Huh? You have 4 points!
 
parallelogram?
 
the krull dimension of a prime ideal is just the height of the prime ideal, right?
 
6:45 PM
Yup.
 
cool. Thanks
 
My yup was to @hansika
I don’t remember that stuff.
 
@TedShifrin okay, earlier i considered 'vectors' $v,w$ and $v+w$, sorry
 
You should be able to prove that you get all the points on the parallelogram and inside.
Now you should finish your question. The shape is very nice.
 
Heyo Ted
 
6:51 PM
Heya Sha!
 
How is it going? Is the temperature unbearingly high too where you live?
 
Hotter than I like, but not crazy. Bad bad drought. The world is about to burn up and dry out …
 
Yea, summers are a painful reminder of where we're heading climate-wise
 
Global warming about 100 years ahead of schedule. Go idiots!
 
@monoidaltransform very much not
if $R$ is an integral domain, then any $R$-module has Krull dimension $dim(R)$
 
6:59 PM
> AVERAGE global temperature. In places like the arctic it’s already warmed double digits. BTW, just a 3-degree average warming is enough to make the world almost unlivable
 
$b\omega+c\omega^2$ lies in the parallelogram with adjacent sides $\frac{-1i}{2}+\frac{\sqrt3j}{2}$ and $\frac{-1i}{2}-\frac{\sqrt3j}{2}$ adding a vector $a$ (i assumed $a$=1 and added with all the four vertices of parallelogram) and joining all the points gave a hexagon. Is it the correct way to do it?
Sorry for late response
 
Yup. You get a regular hexagon. Cool!
 
okay, thank you so much! :)
 
Just take with you the thinking skills you learned!
2
 
yeah!! :)
 
7:20 PM
Excellent advice professor.
👏
 
7:44 PM
@Lukas do you know a slick proof of the following fact? Let $L/K$ be a finite field extension, $A\le\mathrm{End}_K(L)$ a unital $L$-subalgebra and $M$ the set of all $x\in L$ such that multiplication by $x$ is central in $A$. It's easily checked that $M$ is an intermediate extension of $L/K$. Fact: $[L\colon M]=\mathrm{dim}_LA$.
I do have a proof, but it's rather ugly and unenlightening.
 
 
2 hours later…
9:46 PM
Any special relativists in the house?
 
@MoreAnonymous I think that you are looking for the h bar.

 The h Bar

General chat for Physics SE (physics.stackexchange.com). For M...
 
@XanderHenderson i tried there too ...
 

« first day (4391 days earlier)      last day (631 days later) »