« first day (2308 days earlier)      last day (2729 days later) » 

8:14 PM
hi chat
 
good afternoon everybody
 
hi @semi @meow
 
Hi
 
how do projective transformations affect the mapping of $\mathbb{P}^1$ onto $\mathbb{R} \cup \{\infty\}$?
 
$\mathbb{R}\mathbb{P}^1$? (I can't help regardless, but it's worth being clear on that)
 
8:17 PM
if by $\mathbb{RP}$ you mean real projective space, then yes
 
I'm not sure what you mean with your question
 
like consider the mapping $\mathbb{P}^1 \to \mathbb{R} \cup \{\infty\}$ given by "$f(P) = \mathrm{slope of line}$"
 
hi guys
 
i.e. $[x:y]\sim[1:y/x]\mapsto y/x$?
 
yes
i could probably figure this out on my own
 
8:21 PM
I don't know how projective transformations act on $[x:y]$, so I'm no help
(probably it's simple, but that doesn't mean I know it :/ )
 
I think you could and should
 
ok @Alessandro sorry
 
no reason to be sorry, but it's just good for you to spend some time on a question before asking for help
I haven't done any projective geometry in a while, but I'm always here if you are still stuck after thinking about it
 
Wikipedia has some discussion of automorphisms of RP^1 in their page on the latter: en.wikipedia.org/wiki/Real_projective_line
 
automorphisms of the project space
that seems cool
semiclassical
@Semiclassical would you like to check out my project ?
it is about elliptic curve and elliptic curve cryptography.
just want an opinion about it.
 
8:29 PM
Will pass, sorry
 
@meow what do you know about about projective transformation? If you have $S,K$ subspaces of $V=\mathbb{P}^n(\mathbb{R})$ with $\dim S=\dim K=k$ is there always a projective transformation $\phi:V\to V$ such that $\phi(S)=K$? Why or why not?
 
@meow-mix use stereographic projection and iwasawa decomposition
@Alessandro don't projective subspaces correspond to linear ones in affine space?
 
Hello chat
 
Hi @Fargle
 
yes, but that was a question for meow :P @arctic
hi @fargle
 
8:37 PM
oh
 
How's everyone?
 
(also @meow see KCd's blurb on Iwasawa decomposition for SL(2,R), it essentially carries over to PGL(2,R))
aight
 
where can I read about this Iwasawa decomposition @arctic? I've never heard the term before but it seems interesting
 
(most discussions of iwasawa decomposition do so in full generality, which is pretty advanced lie theory)
 
yeah, wiki starts talking about lie theory right way and I know nothing about it
 
8:40 PM
hi @arctictern
 
that pdf on the other hand seems very accessible, I added it to the list of things I should read, which means I'll remember to do so in a few months maybe...
 
that is nice accessible paper
I will read it after semester is over.
 
@Alessandro it's really cool doing it for GL(2,C) and classifying effects on the sphere under stereographic projection (see the youtube video "Mobius transformations revealed")
 
@arctictern would you like to check my elliptic curve project ?
 
I also have a few answers which cover this stuff
I know next to nothing about elliptic curves
 
8:42 PM
oh
 
you'll regret this when I start reading it and pester you with questions :P @arctic
 
no I won't
 
@arctictern Holy cow, that's a nice result.
 
@arctictern have you taken homological algebra course before ?
 
nope
@Fargle like I said, cooler with PGL(2,C) and stereographic projection
actually you can do it over quaternions too
 
8:46 PM
I wasn't serious, but it's always awesome to see how willing to help people here are
 
@Fargle hi
@Alessandro I learned more math here than university.
 
@arctictern I just like how visually intuitive the proof seems to be. Rotate $ge_1$ clockwise onto the x-axis, scale $ge_1$ by $1/r$ (which requires scaling $ge_2$ by $r$ to have det 1), and shear $r(ge_2)$ back to $(0,1)$.
 
Hello, what are you talking about?
 
@Xammm math.uconn.edu/~kconrad/blurbs/grouptheory/SL(2,R).pdf This link, on Iwasawa decomposition.
 
@Fargle seems interesting, but sadly I don't know anything about Iwasawa decompostion :c
 
8:54 PM
I didn't either before reading about it. It's fascinating.
 
Very useful in geometric group theory
 
It's about Lie groups, right? Hope someday I'll learn Lie algebras, Lie groups and similar stuff.
 
yeah @Xammm
 
Good night/morning all chat!
If some user is interested in Analytic Number Theory they can see/study from the class of professor Ram Murty in YouTube from the official channel matsciencechannel, the name of the course is Introduction to Analytic Number Theory
 
9:09 PM
Oh, thank you @user243301 !!!
 
Good morning/nights all users. Bye, bye.
 
Bye
 
Thanks to you @Xammm is a great professor. The course is of the more high quality, is incredible.
Bye alls.
 
9:22 PM
hey @Fargle
 
Howdy @Semi, how goes it?
 
eh, tired. mid-late-afternoon here
 
Same here.
 
This is the time when my brain doesn't do much that's useful
 
I'm basically only capable of doing things in the evening.
 
9:28 PM
when I tell engineers I like math they picture me doing stuff like differential equations
 
Has anyone seen @robjohn recently? Has he transitioned into the grinch who stole Christmas, yet? Of do I need to wait a few days? Hmmm...I wonder how he'd look in green!
 
@Sophie A better conception than the layman's...
 
What kind of math do you tend to do, out of curiousity? @Sophie
 
I like number theory a lot
 
Ah. That's a bit different than DE's, yeah.
 
9:37 PM
I think I know too little to have other consolidated interests yet
 
9:49 PM
Is someone here familiar with the Lagrangian formulation of classical mechanics by any chance?
 
ask semi c
 
@Alessandro Yup (ask away, if you were waiting to)
 
I have a lot of general confusion that I'm trying to organize into sensible questions
 
OK
@MikeMiller So I think I just about proved that the Euler class is PD to generic zero locus. Now, do you know any nice examples of bundles with nonzero Euler class that are easy to provide using this? So far, I listed a bunch of non-zero Euler characteristic spaces, but I'd love to be able to use e.g. normal bundles for some examples.
 
10:06 PM
yeah I am also familiar with lagrangian formulation @Alessandro
 
I also don't have my notes available right now, if it doesn't bother you I'll ping you sometimes soon-ish so that you can look at my question when you have time, I'm not in a hurry
 
sure
 
@Danu Normal bundle of P^1 in P^2.
 
thanks a lot
 
@BalarkaSen Right, neat.
 
10:08 PM
"if polynomials of degree n are equal on n+1 values then the polynomials are the same for all values" <- can someone say what i need to know to prove this?
 
@Danu Circle bundles over $S^2$.
rather complex line bundles; sorry
 
of which my example is a special case, so that's better
 
@MikeMiller So you mean $\Bbb CP^2\# \overline{\Bbb CP^2}$?
 
just any complex line bundle on S^2. those are classified by the euler class.
 
@Null That $\Bbb R_n[X]$ is vector space of dimension $n+1$ @Null
 
10:10 PM
@Danu I don't mean $\Bbb P^1$ bundles, I mean complex line bundles
 
@BalarkaSen My audience ain't comfortable with sheaf cohomology and I don't know a 3-second way to prove this otherwise
 
You can prove that easily if you know something about polynomial interpolation, but there are probably plenty of ways to show so @null
 
@MikeMiller Yeah, I don't know why I thought about that, sorry :P
 
@Danu Clutching functions
 
actually I like @astyx's suggestion much more than my own
 
10:10 PM
Bundles with structure group $G$ over $S^n$ are classified by $\pi_{n-1} G$
 
@MikeMiller Definitely not a 3-second thing for my audience
 
yeah fine
 
thanks @Alessandro @Astyx :)
 
You're welcome
 
^
(was ^'ing to clutching functions, internet delayed it)
 
10:12 PM
which gives a nice corollary: $G$-bundles over $S^3$ are trivial
 
Very nice
So any $4n$-manifold with intersection form with non-zero diagonal entries will give normal bundle examples right
 
sure though you have to convince the audience those can be represented by submanifolds
 
yep
 
actually they probably can't in general, rip rip rip
 
can it always? I only know it can for n = 1
 
10:14 PM
But for 4-manifolds they can right
 
yeah
 
So what goes wrong in higher dimension?
 
whenever $n=1$ or the manifold is $(2n-1)$-connected you're fine
what goes right?
 
there's a 7-dim example by Steenrod.
iirc
 
Idk, surfaces are nice for $n=1$? :P
 
10:16 PM
you can always represent a codimension 2 homology class by a submanifold or a homology class of dimension up to, uh, 4? 5?
 
I have no idea what happens when you try to resolve self-intersections for immersed 4-manifolds
 
link of a generic intersection looks like a hopf link i guess. so you remove stuff and glue a siefert surface to it
 
Of what?
I was thinking about 4-manifolds in an 8-manifold
 
any two surfaces in a 4-manifold
oh
 
surfaces in a 4-manifold only intersect in points (transversely)?
I should shut up
 
10:18 PM
you can always set things up so that an immersed manifold intersects itself transversely
 
I don't know what a Hopf link is :P
 
It's just a version of the transversality theorem for immersions.
@Danu Two circles linked in the simplest possible but nontrivial way.
i.e. linking number 1
 
In what sense does a self-intersection of a surface look like that?
 
the problem is you think homology classes in higher dimensions come from immersed manifolds at all
@Danu It looks like two transverse planes in $\Bbb R^4$
 
@MikeMiller I saw that picture---but I don't see how that's a Hopf link :P
 
10:20 PM
When you intersect that with the unit sphere you get the Hopf link
 
Ah.
 
huh, do you? Maybe only if it looks like two planes in $\Bbb C^2$, I'm not sure if two generic planes in $\Bbb R^4$ look like that
 
Ah, I see. Cool
 
but whatever
Yeah i guess it does always look like that.
 
I was thinking about $\Bbb C^2$ anyways :P
So, any other nice examples of non-trivial self-intersection of submanifolds?
(other than connected sums of CP^2 and its opposite orientation partner)
 
10:21 PM
You can just perturb the two 2-planes to be complex lines, I guess.
 
there's an element of GL4 that takes any pair of transverse planes to any other pair
 
ah, yeah
 
@Danu I mean, most 4-manifolds
 
@MikeMiller Yeah? In what I've seen in my course we didn't do so many with diagonal entries.
 
Hello!
 
10:23 PM
diagonal is zero in SC 4-manifold iff it's homeo to $\#_n S^2 \times S^2$
 
Then again, we didn't do many.
 
Hi @SimpleArt
 
diagonal in S^2 x S^2 has nontrivial self intersection
 
the diagonal of the intersection form, you twerp
 
hehe
 
10:24 PM
If anyone has two seconds, this answer I posted here needs one more vote to reopen. Don't just cast a vote, but actually take a moment to look at it please.
2
Q: How do you prove this very different method for evaluating $\sum_{k=1}^nk^p$

Simple Art I found the following formula in my previous question. This differs from my previous question in that I want an alternative proof of the below recursive formula for calculating $\displaystyle\sum_{k=1}^nk^p$. Suppose I had a function recursively defined as $$f(x,p)=a_px+p\int_0^xf(t,p-1)...

 
I agree it's not a dupe.
 
@MikeMiller which class you think I should take from the following three ? (1)Algebraic topology (2)Functional analysis 2 (3)Representation theory of lie groups.
 
@MikeMiller I was giving an example to Danu's self-intersection question. Didn't want to sound I was trying to contradict you; internet just keeps making bad timing.
But it actually looks quite funny now
 
@MikeMiller Thank you for taking the moment to look at it. Cheers
 
@Adeek It depends on what you like the most. I like algebraic topology, I use functional analysis, I wish I knew the representation theory of Lie groups.
 
10:27 PM
haha
Lie algebras are nice
(i think)
 
maybe I should attend all of them and see which one I like the most
 
Not a bad idea.
 
I could audit the one I don't like and I am not taking
 
Agreed
I'd drop FA without a second thought :D
 
functional analysis sounds good stuff
 
10:28 PM
XD
 
FA is very interesting @Danu
 
I know
 
I use them in pretty much decreasing listed order
 
I don't know anything about it tho
 
But the other two much more so IMO
FA is probably the most universally useful though.
Can't really get around it, it seems.
 
10:29 PM
what's FA about?
 
infinite dimensional vector spaces
 
right
@MikeMiller Weren't you talking about that today?
 
was I?
 
This is the outline for FA
Locally convex spaces, weak topologies and duality in Banach spaces, weak compactness in Banach spaces, structure of classical Banach spaces, local structures, infinite-dimensional geometry of Banach spaces and applications.
 
cool
 
10:30 PM
that sounds interesting, guess I'll take FA next year (also because it's not mandatory but kinda strongly suggested)
 
not the sort of functional analysis I use
leads more into operator algebras and the modern study of functional analysis, probably
 
@SimpleArt do you know the closed form of Faulhaber's formula with Bernoulli numbers?
 
yeah we have here like 5 functional analysis classes. Here Algebra and analysis is quite strong @MikeMiller
 
What are applications of Lie algebrae ?
 
@Sophie Don't know the Bernoulli numbers very well. Now that you mention it, I could try some induction
Sounds very tedious though.
 
10:32 PM
In my last day on a complex analysis "set of informal lectures" with a person, he gave a smallish introduction to abstract Fourier theory. we started from measure theory and ended with infinite-dimensional representation theory of groups.
 
In mathematics, Faulhaber's formula, named after Johann Faulhaber, expresses the sum of the p-th powers of the first n positive integers ∑ k = 1 n k p = 1 p + 2 p + 3 p + ⋯ ...
 
we did some abstract fourier theory in our class that I am taking
very nice.
 
@Sophie Thx, I'll brb with that
 
@Astyx Physics
 
10:33 PM
If you want/manage to do so, feel free to post it as an answer
 
@Danu What field of physics ?
And how ? (succintly)
 
@SimpleArt if I find a proof I'll do
 
@Astyx High energy particle physics
All the particles predicted between 60's and 80's came from representation theory, with the exception of the Higgs.
 
That seems quite cool
 
Once the physicists realized you're supposed to use Lie algebras/groups it was all very "easy"
Pretty awesome time that must've been.
 
10:36 PM
And how exactly does one use Lie algebra/groups ?
 
I don't have time right now---maybe some other time ;)
 
Ok, thanks a lot :)
Any way I'm gonna go to bed, good day/night to all of you
See ya
 
But essentially one says that certain sets of particles that are physically related span a vector space on which the relevant Lie algebra acts (i.e. form a representation space)
 
Before I leave : would you happen to have any reference to a good introduction to Lie algebra ?
 
@Sophie Ugh, I've never used Bernoulli numbers before, so this is somewhat difficult for me.
Plus the notation on en.wikipedia is bad.
 
10:40 PM
@SimpleArt what's wrong with it?
 
@Sophie I don't know what to do with them or any formulas relating to them
 
Bernoulli numbers are just the numbers on the Taylor series of $\frac{x}{e^x-1}$
 
Yes, and they can also be defined as the solution to some crazy contour integral (derived from that above), but I feel like I'm somewhere I don't yet belong is all. Don't worry too much, I'm sure I'll figure this out
 
I think "and they can also be defined as the solution to some crazy contour integral" applies to any constant ever
 
@Sophie I was talking about the coefficients of the Taylor of your function ^. The coefficients of any Laurent series are representable by Cauchy's nth derivative formula
Just a fun little fact there.
But uh...
How do I simplify the following?
$$\frac p{p+1}\sum_{k=0}^p(-1)^k\binom{p+1}kB_k\frac{n^{p+2-k}}{p+2-k}$$
 
10:57 PM
is there a fast way to check that $[Q(\sqrt(2),\sqrt(3),\sqrt(5)) : Q(\sqrt(2),\sqrt(3)) ] = 2$ without checking if $\sqrt(5) \in Q(\sqrt(2),\sqrt(3),\sqrt(5))$
 
is recursion with real values stupid? Like $f(x)=f(x-1)\cdot 2$
 
@Null Huh? I see that $f(x)=2^xf(0)$
 
what do you mean @Null
 
11:48 PM
@SimpleArt only for $x\in\mathbb{Z}^{\ge0}$
 
11:59 PM
@DHMO someone gave a proof of that diophantine equation I had mentioned
 

« first day (2308 days earlier)      last day (2729 days later) »