« first day (2284 days earlier)      last day (3031 days later) » 
00:00 - 21:0021:00 - 00:00

00:01
@DHMO Yep sorry I am wrong
I found my notes
I will just say this:
To show that a statement about universal truth (one with) is true, we need to go through every single element in our consideration and confirm that each element satisfies the requirement(s) set by the statement.
To show it is false, we only need to provide an example that doesn’t satisfy at least one of the requirements set by the statement. Such an example is called a counterexample.
To show that a statement about existence (one with) is true, all we need is to find an element (an example) that satisfies the requirement(s) set by the statement.
To show it is false, we need to go through every single element in our consideration and confirm that no element satisfies all the requirements set by the statement.
@DHMO dunno. I'm just trying to translate the kind of strict ordering $x_1<x_2$ that $\mathbb{R}$ has to $\mathbb{C}$, but it doesn't seem possible
@DHMO ...I suppose, in the sense that $\mathbb{R}$ is apparently an ordered field but $\mathbb{C}$ is not
@DHMO good morning
I have a question in pair of straight line
I feel much more comfortable with a statement saying: "Given a group $G$, $\exists$ $g \in G$ $s.t.$ (etc etc)"
I should have said that they are not easily layerable/stackable
yeah
If we have two equations of pair of lines (say s_1 and s_2) and both has a line in common,then how to find that common line?(these are homogeneous equation)
00:12
$\forall y\in Y:\exists y=y$ would that even be correct?
@saturatedexpo $\forall y\in Y,\exists y,\; y=y$ is more sensible but still funny
00:56
would it be ok if i took both algebraic topology and algebraic geometry? i want to get a feel for both
01:28
Super quick question about PDEs if anyone is able to help: math.stackexchange.com/questions/1998473/…
if i have to show that some (G,+,$\cdot$) is a commutative Ring, do i only have to show that ab=ba? because that stands on wikipedia
-.-
@saturatedexpo well if you show that $ab=ba$
OR do i have to show that $a\cdot(b+c)=a\cdot b+a\cdot c$
thats given because $G$ is a ring
mmh, in my assignment not really
01:33
anyways
if you show that $ab=ba$
then for $b,c \in G$
the following holds:
$a\cdot (b+c) = (b+c)\cdot a$
and then, by distributivity
$ab+ac=ba+ca$
since $ab=ba$, we subtract that equation from this to get
mmh, so i wrongly proven a(b+c)=ab+ac, instead of the much much easier one. But does this then still be a prove for commutativity?
(i mean not wrongly, just unneccessary)
oh
i just realized what you meant
if you need to show $G$ is a ring then you have to prove the ring axioms lol
i mean its unclear to me what i have to show
$(G,+,\cdot)$ is given and also + and $\cdot$.
ok
then prove the ring axioms lol
and in addition i have to show that ab=ba? (for "commutative")
01:37
youre correct
i feel retarded now
but im fine with that
sometimes you just have to do dumb things :D
@meow-mix and +, and $\cdot$ are just placeholders and could be replaced by lets say - and $\circ$ ?
$+$ and $\cdot$ are binary operations
they are mappings $G \times G \to G$
user228700
Hi everyone :-) I'm having a little trouble deriving all the relations and values of constants related to the standard ellipse, $x^2/a^2+y^2/b^2=1$.
@meow-mix thanks, definitions help
i think i got it now
alright, im glad i could help
user228700
01:44
I have some small questions...
@Kaumudi "Just ask; don't ask to ask."
user228700
I derived the equation $x^2/a^2+y^2/b^2=1$ by assuming the fixed sum of distances of any point on the curve from the two fixed points to be $2a$ and the coordinates of the focii to be $(-c,0)$ and $(c,0)$.
user228700
But now, I'm looking to derive the value of $e$, the eccentricity of the ellipse. $e$ is defined as the ratio of the distance of any point on the conic from a fixed point, the focus (in this case, focii) to its distanced from a fixed line, the directrix.
user228700
And I'm having trouble with this because although I know the coordinates of the focii, I don't know the equation of the directrices.
user228700
02:01
My book has just assumed the focii to be $(ae,0)$ and $(-ae, 0)$ and the directrices to be $x±a/e=0$
user228700
Is this OK?
user228700
OK, I've boiled it down to one small question. How is it that the distance b/w the point $(0,b)$ to the focii $(±c,0)$ is $a$ in the following diagram:
user228700
user228700
(Pls excuse my drawing skills...its lack thereof :-P)
user228700
02:15
@DHMO
@Secret nice secret you shared with us. pun intended.
Well, I actually like pure maths as much as applied. To me pure maths is the study of the mathematical objects themselves and I do have a rather physical projection on mathematics in general, going as far to treat all mathematical objects as something you can fiddle with
Just my abstract algebra experiments, in particular division by zero algebra is basically pure maths
my dream is to make as much pure math to applied math as i can in my lifetime. even if that will only be one big theorem.
but i can live with it, that some do it just for the puzzles :)
is there some tool that notates all set operations which are seminested to the extendeds way possible? If not i might write my snippet.
oh, i mean A\B and $A\cap B$ and $A\cup B$, all with brackets
like $A\cap (B\cup C))$
into $A\cap B\cup A\cap C$
02:45
I am not sure what you mean, but both $\cup$ and $\cap$ distributes over each other
yes
but doing it over this: $\Leftrightarrow x\in ((A \cup (B \cup C) \backslash (B \cap C)) \backslash (A \cap (B \cup C) \backslash (B \cap C)))$
is PITA
 
1 hour later…
03:49
3
Q: Nature of $G$ when $N$ is cyclic, normal subgroup of $G$ and $G/N$ is cyclic

Jessy CatLet $N$ be a normal subgroup of $G$ and both groups $N$ and $G/N$ are cyclic. I need to prove that $G$ is generated by at most two elements. To that effect, what sorts of things do we know about $G$ if $N$ is a subgroup of $G$ and $N$ and $G/N$ are both cyclic? I know that all cyclic groups a...

I've got a 100 point bounty on this question.
@Danu Yes, it does.
The point is the cohomology ring of the disjoint union is natural isomorphic to the direct sum of the connected components.
The map being, (i^*_1, i^*_2, ..)
i_1, i_2, etc are inclusions.
 
1 hour later…
05:01
Anybody in here know Alan?
05:19
me :)
ah no
no i dont know him
user228700
06:17
Hi again. I've got a quick question.
user228700
The equation of tangent to the ellipse $x^2/a^2+y^2/b^2=1$ is given by $y=mx ± \sqrt{a^2m^2+b^2}$
user228700
Where $m$ is the slope of the tangent. Clearly, for every slope, there are two such tangents, with different y-intercepts.
user228700
These two tangents are clearly parallel to each other.
user228700
My textbook has given a result related to this and it starts off with "Let this tangent pass through a point $(h,k)$"
user228700
My textbook has then gone on and written $k=mh ± \sqrt{a^2m^2+b^2}$ by substituting this point on the equation of both tangents!
user228700
06:24
How is it that these two parallel tangents will pass through a common point $(h,k)$?!
user228700
Or have I misunderstood? 'Cause my textbook says, at the end "If we know what $(h,k)$ is, we can solve the obtained quadratic equation for two values of $m$ "
user228700
Pls. help 🙏
user228700
07:18
Never mind, never mind, I figured it out!
@Kaumudi What was it?
user228700
07:36
in The h Bar, 27 mins ago, by Kaumudi
U know, maybe I am right, in that the two tangents don't intersect.
user228700
If u click on that link, you will be redirected to The h Bar, where you can see how I figured it out :-)
user228700
Wasn't at all the big deal my brain had thought it to be-my brain is sometimes very dumb, that's all :-P
08:03
@Kaumudi try finding the sum of the distances from the two foci to (-a,0).
08:44
hi. I'm referring to the correct answer of this : math.stackexchange.com/questions/265396/…
could some body please explain why that function is one to one
09:40
So I've been sent some papers by a professor who wants me to do research, but the content of the papers is above my level of expertise. How would the denizens of the chat recommend I read papers like this?
@Fargle start reading textbooks on the background
@TobiasKildetoft Thanks!
10:04
Hey guys I need help
I need to derive an equation relating x to k from the equations $3x+2y=5$ and $3x^2+2y^2=k$
In the first equation, solve for $y = \dots$ and then substitute that $y$ in the second equation?
\dots?
So then I get $3x^2+2(5-3x/2)^2=k$
Correct?
@SylentNyte It is correct if what you mean is that $3x^2 + 2\left( \dfrac{5-3x} 2 \right)^2 = k$
10:25
Yes
10:55
@Fargle What's the paper on?
@BalarkaSen Solutions to second order systems of the form $q''(t) + Mq(t) = f(q(t))$ where $M$ is symmetric and positive semi-definite, and where $f(q) = -\nabla U(q)$ for some smooth function $U$.
The paper starts off rather quickly with "well obviously this is a Hamiltonian system with $H(p,q) = \frac{1}{2}(p^Tp + q^TMq) + U(q)$" and I'm like "I think you need to rethink the use of the word 'obviously'"
Yeah, I don't know anything about this either. You should consult a textbook regarding this.
Unfortunately, my library is thin on books concerning systems of ODEs.
The professor who suggested this paper might be able to recommend you some, perhaps.
Hopefully--I'll consult him about it.
He's just, like, "You could write this paper in like a month! I'm sure of it!" I fear he drastically overestimates my content knowledge, haha.
11:02
Also if it's only 1 line you can't understand, you probably shouldn't go down the rabbit hole trying to understand it. Try understanding the big picture first, and then figure stuff out.
Oh, no, it's the whole thing. I can't see the forest.
Fair enough.
Appreciate the perspective though, as always.
Man, Aluffi is a weird book.
Chapter 0?
Yeah.
Just, the structure of it. I like it--it's just so very different from other algebra texts I've looked through.
11:15
I don't dislike it, but I think one should know the concrete story before the abstract one presented in Aluffi.
Agreed.
Once done that, Aluffi's way would be much more enlightening and actually exciting.
I'm just stumbling a teeny bit with categories.
I discourage pondering on category theory too much. I picked it up on the way; I think most people who doesn't study it pick up the necessities that way.
Why do you discourage it?
11:25
Because the fundamentals on category theory are a language for packaging a bunch of stuff (eg group/ring/field theory and study of their isomorphism types, topological spaces upto homeomorpism, smooth manifolds upto diffeomorphisms, algebraic varieties upto variety-isomorphisms etc). It comes after those, not before.
Otherwise no motivation is supplied for learning the abstract terminology of category theory.
I suppose that's true. Stuff like the isomorphism theorems fall directly out of the category theoretic language, but may appear alien to someone who hasn't seen specific examples of said theorems.
Right.
By discourage, I of course meant discourage for the beginner. The categorical language is sometimes useful for making connections between vastly different objects of study, and sometimes actually useful for proving concrete things - the thing is it bares stuff out of the unnecessary data so you know what you want to see.
I myself have found it useful - to my surprise - in proving something very pictorial, in topology.
Haha, I guess my main trouble is knowing whether to categorize myself as a beginner or not.
...no pun intended.
Are you studying algebra for the first time?
No, but I'm weak on a lot of the later results.
e.g. Sylow theorems, later things with ideals, extensions, Galois theory.
11:32
Then you should study it all concretely. I recommend Artin.
Once done that, you can get back to Aluffi. I like his unified presentation of Sylow.
Good call. I suppose this approach would only cause me more headaches.
Yep. The thing is if you go through Artin you'd get a LOT examples. Then Aluffi would tell you how to package it. So you'd have an all-round, concrete, understanding of algebra :)
Hooray!
Is there an analogous book for analysis, or am I already on the right track with baby Rudin for that?
Artins biography on wikipedia is written in an entertaining manner
> Artin had recourse to an encyclopedia, which he once consulted for help in dealing with the cockroaches that infested the Austrian barracks. At some length, the article described a variety of technical methods, concluding finally with—Artin laughingly recalled in later years—“la caccia diretta" ("the direct hunt"). Indeed, “la caccia diretta” was the straightforward method he and his fellow infantrymen adopted.

Artin survived both war and vermin on the Italian front, and returned late in 1918 to the University of Vienna, where he remained through Easter of the following year.
Heh.
11:40
Hehe
So $\cos^{-1}x = \ln(-iz+\sqrt{-z^2-1})$ ?
@Fargle Real analysis?
@BalarkaSen Yes, sorry.
Not that I am aware of. I never really systematically studied real analysis.
I like the book by Dieudonne: Foundations of Modern Analysis
but I don't know what kind of analysis book you are looking for, so maybe this is not appropriate
@DHMO the formula is a little different, but you can find the correct thing by taking derivatives of both sides (the derivative of $\cos^{-1}$ can be found from the inverse function theorem) and then getting the right integration constant
11:55
@s.harp ok, thanks
@s.harp I thought that was mostly an encyclopedia, but no idea. I once looked at it for a technical generalization of Clairaut's theorem about mixed partials.
Ew, Artin defines $\mathbf{i} \in Q_8$ as $\left[ \begin{smallmatrix} i & 0 \\ 0 & -i \end{smallmatrix} \right]$.
@Balarka Dieudonne has two "books" on analysis, one is this, which is a textbook and the other is "Éléments d'analyse" which is 9 books that I think were never finished
Oh. I don't remember which one I am talking about.
It still works, but I'm just so used to $\mathbf{i} = \left[\begin{smallmatrix}0&1\\-1&0\end{smallmatrix}\right]$...
11:58
@saturatedexpo To prove: $\forall q\in\mathbb{N}:\exists n,m\in\mathbb{N}_0:q=f(n,m)=\frac{(m+n)(m+n+1)}{2}+m+1$
Basecase ($q=1$): We find $n=m=0$, such that $f(0,0)=\frac{(0+0)(0+0+1)}{2}+0+1=0+1=1=q$
Induction step: $q\to q+1$
$q+1=(\frac{(m+n)(m+n+1)}{2}+m+1)+1=\left\{\begin{array}{ll}\frac{(m+1+n-1)(m+1+n-1+1)}{2}+m+2=f(m+1,n-1)&\mbox{, if }n>0\\\frac{(m+0)(m+0+1)}{2}+m+2=\frac{(m+1)(m+2)}{2}=f(m+1,0)=f(m+1,n)&\mbox{, if }n=0\end{array}\right.$
Thus, we can find for any arbitrary $q$ at least one pair $(n, m)$ which proves surjectivity.
@Fargle But you need $j$ or $k$ to be imaginary-entried matrices in any case don't you?
@NaCl the problem is, how did you find out the algorithm?
@DHMO what algorithm?
> Dieudonné is putting together a monumental treatise which will comprise d volumes; the best estimate we know on d is 8 < d < 12, but the upper limit is not certain
12:01
lol
@NaCl the induction step
@BalarkaSen Yes--I just like the version I posed because it falls naturally out of the isomorphism between $\Bbb C = \{a + bi|a,b \in \Bbb R\}$ and $S = \{ \left[ \begin{smallmatrix} a&b\\-b&a \end{smallmatrix} \right] | a,b \in \Bbb R \}$.
maybe you need \{ and \}?
@DHMO because $f:\mathbb{N}\to\mathbb{N}_0\times\mathbb{N}_0$ and thus I need to have an induction step of $q\to q+1$ to prove for all $\mathbb{N}$
I see your point.
12:03
@NaCl I know how your proof works. My question is, how you managed to come up with that step.
I suppose there's a mnemonic advantage to his formulation though.
The $\mathbf{i}$ matrix is the one with $\pm i$ on the diagonal.
@DHMO You could ask that to anyone who comes up with any type of proof. For $f$ to be surjective, $\forall q\in\mathbb{N}:\exists m,n\in\mathbb{N}_0:q=f(m,n)$
And then I just thought: Why not induction? As that is one of our topics anyways
I have to decide what I want to study.
@Balarka: what are you considering?
12:09
differential geometry of surfaces, connections, characteristic classes, a topology problem, schoolwork, procrastination.
differential geometry of surface you mean erlangen programm geometric structure things?
@s.harp not sure what you have in mind by that. i just means study of surfaces immersed in R^3: it's first/second fundamental form, curvature, etc
I meant something where you have a model space $X$ and a group of symmetries $G$ acting on it and then a $(X,G)$ manifold is a manifold with charts in $X$ so that the chart switch maps are equal to the restrictions of elements of $g$ onto the connected components
this kind of thing allows you (for example if you take $X=\mathbb R^n$ and $G$ affine transformations) to talk about lines in your space or ratio of lengths of points on lines etc without needing to go the canonical route of having a full riemannian metric
sure. not what i meant
in that case my recommendation from nowhere is to say characterstic classes if there is daylight and schoolwork if it is nighttime
12:21
well, darn. schoolwork it is then! :)
13:02
Do local isometries of Riemannian manifolds preserve the properties "to have positive injectivity radius" and "to have bounded sectional curvature?
13:44
@NaCl perhaps $$
\begin{align}
\sum^{n+1}_{k=1}{a_kb_k} &= a_{n+1}b_{n+1}+\sum^{n}_{k=1}{a_kb_k} \\
&= a_{n+1}b_{n+1}+A_nb_{n+1}+\sum^{n}_{k=1}{A_k(b_k-b_{k+1})} \\
&= a_{n+1}b_{n+1}+A_nb_{n+1} - A_{n+1} (b_{n+1}-b_{n+2})+\sum^{n+1}_{k=1}{A_k(b_k-b_{k+1})} \\
&= A_{n+1}b_{n+2} + a_{n+1}b_{n+1} - (A_{n+1} - A_n)b_{n+1}+\sum^{n+1}_{k=1}{A_k(b_k-b_{k+1})} \\
&= A_{n+1}b_{n+2} + a_{n+1}b_{n+1} - a_{n+1}b_{n+1}+\sum^{n+1}_{k=1}{A_k(b_k-b_{k+1})} \\
&= A_{n+1}b_{n+2} + \sum^{n+1}_{k=1}{A_k(b_k-b_{k+1})}
@AntonioVargas woah, good catch
@AntonioVargas Why don't you post it as answer? :)
14:43
Is following true? The only interesting prime-generating polynomials are those, whose constant term is 1 because otherwise you just plug in the constant term and the result is divisible by it?
@Adam That's correct.
More generally I think it's true that there are no polynomials which always generate primes.
@BalarkaSen No, those do exist, they just take several variables
Well, with integer coefficients, I meant.
@TobiasKildetoft Ah, ok.
I had one-variable ones in mind, but still interesting.
The known ones have quite large numbers of variables or large degrees
and by these I mean ones that not only generate primes, but generate all primes as well
I know. I need it as an example of the why we need proofs - end Eulers polynomial seemed too trivial as you can subtitute the constant term.
Can anyone point me in direction of a prime-generating polynomial with 1 as a constant term?
14:51
@Adam Do you want one that actually generates just primes, or what?
No, just many primes?
@RonGordon That's a thought-provoking article
Interesting yet also alarming
15:14
hi chat
Hello @Semiclassical
Hey @Semiclassical
@AntonioVargas heya
Howdy @Semi
@Adam 6n+1
15:19
interesting...
Call me dumb, but isnt it one of those that produce infinitely many primes?
@teadawg1337 I thought so too. Extrapolate to our little world here and the pattern more or less fits. I have no idea what to do though - it does get discussed every now and then.
@Adam Yep. It generates infinitely many primes by Dirichlet's theorem
15:50
To make the question more specific: are there any polynomials over $\Bbb Z$ that generate infinitely many primes, and only primes? Or is there a no-go theorem for that? Or @Tobias, did you mean that the ones of which you spoke do in fact do this?
Oh: there is a no-go theorem for that.
16:12
See discrete mathematics for computer science by Lehman for an easy proof.
I saw a pretty easy one.
where?
Let $P(n)$ be the polynomial; let $P(1) = p$ so that $P(1) \equiv 0 \mod p$; but then $P(1 + kp) \equiv 0 \mod p$ for all $k$, so that $P(1 + kp) = p$ for all $k$, requiring $P$ constant.
Wikipedia.
@NaCl nice. thanks
16:46
As there's no discussion, I don't feel bad about derailing any discussions, but: what music do people in here like to do math to?
I'm starting to wonder if my latest question was better suited for MO... It's stumped me for months, but I posted it on MSE because I'm under the impression that integral questions aren't taken seriously by professional mathematicians. I just seem to be falling victim to homework questions and basic questions. It also makes me wonder if my question is as significant as I believe it is
@Fargle I usually focus better without music
would have to be something generic and not too distracting. so silence for me @Fargle
I guess that's fair.
@Fargle pink floyd, radiohead, various game OSTs. vocals can't be too pronounced or I get distracted
I like to have something quiet but a teeny bit engaging in the background. If it's something I already know it's more likely that I'll sort of fade it into the background.
@MikeMiller All good choices.
16:48
Excellent choices
What that means for me usually is instrumental prog: Scale the Summit, Plini, Animals as Leaders, and the like
Radiohead is amazing
@MikeMiller I always liked the soundtrack to Journey (the video game)
One of my favorite game soundtracks has to be VVVVVV.
i like the original tetris soundtrack
16:50
@BalarkaSen Inspired by trad. Russian folk, right?
...somehow I feel like you're the one who told me that.
far more obscurely, Tetrisphere (for the N64) had a great soundtrack
Rez had good tracks as well
Really? I don't think I ever mentioned the Korobeiniki.
I always loved Earthbound's soundtrack, but then I found out a lot of it was (lovingly) ripped from The Beatles, and it all made sense because The Beatles are one of my favorite bands, period.
@BalarkaSen This may be a case of déjà vu.
16:52
If we're going to mention Earthbound, gotta mention Undertale
TRU
That OST is...inspired.
yeah.
Not sure I can pick a fav track from that. depends on my mood
Morrowind's soundtrack is great too.
I rather like the (to borrow a TVTropes phrase) fridge brilliance of the main beat of the main theme being a heartbeat, given the final boss.
Hans Zimmer's movie soundtracks/compositions are nice, but I have never tried doing math on those.
hi, pardon me, what kind of foundation to learn things such transformation like fourier transform, laplacian transform. I try to learn math again but all I read is transformation like scaling, translate, rotation. Thanks
17:02
You're interested in integral transforms, not transformations. somewhat different terminology
woah thank you very much
Suppose $A,B$ are monoids and let $A\amalg B$ be their coproduct in the category of monoids, comprised of reduced words. Does the relation $ab=ba$ for $a\in A,b\in B$ imply either $A$ or $B$ is zero? Seems so since there are no relations between $A,B$ in the coproduct, but I'd like to make sure..
but I'm not sure what kind of transformation I need
Sort of obvious, but have you read Wikipedia's page on integral transforms? en.wikipedia.org/wiki/Integral_transform
The foundation, broadly speaking, is integral calculus
thank you very much @Semiclassical, I'm developing computer vision software to analyse skin losion, but fatally, very weak mathematical background which resulting a very bad engineer like me
17:05
mmkay
@Fargle try listening to the sword and sworcery EP
@MikeMiller Will do. What's it like?
You'll probably want to grab a book on mathematical physics.
I mean skin lesion, not lotion, lol, alot of geometrical things on the pic pixels
Good friday, all users!
17:09
To get the physical intuition for the Fourier transform, suppose I have surface with a narrow slit @Hey-men-whatsup
Hi guys, what is the name for a generalised version of symplectic matrices such that $M^{-1}\Omega M=\Omega$?
"matrices $\Omega$ commuting with $M$"?
Woop-dee-woop, my seminar talk went great :)
@Fargle Emancipator
I can shine a laser on that, and that'll amount to hitting that surface with a specific wavelength of light
@Fargle proggish
17:12
@MikeMiller aww yiss
More explicitly, is there (inset name for matrix) $M$ such that $\Omega$ is "self similar matrices"
self-conjugate, perhaps
I can describe the illumination of that as a step function, e.g. $f(x)=1$ if $|x|<1$ and $f(x)=0$ otherwise. if the point is within the slit, the laser reaches it; if not, it doesn't.
Hey @MikeMiller how are you doing
Now, suppose I put a plate after that slit; I'll see an intensity pattern of the laser light shining through it. (i've actually done this in intro physics labs)
@Semiclassical, sorry again, do have any recommendation of books to learn this kind of integral transformation, I'm not sure if found it in my calculus book that I've just purchase
17:15
I like Arfken.
that pattern, as a function of position on the plate, is essentially the Fourier transform $F(k)$ of that profile function $f(x)$
Ok thank you very much again Semiclassic
np
If this link had working pictures, I'd probably suggest you use it: w.astro.berkeley.edu/~jrg/ngst/fft/optics.html
this is the stuff behind our motivation/ projects link.springer.com/chapter/10.1007%2F978-3-319-08491-6_16
17:23
@Danu I'm good. Sounds like you are too!
@MikeMiller Yeah, I'm pretty happy about it.
I have been sick/sick-ish all week so I'm happy I managed to pull through on this.
@Hey-men-whatsup Not seeing a lot of detail in there. All I see is that one can look at the 2D Fourier transforms of those pictures; there's really nothing in there showing that said transforms provide a useful characterization.
'showing', in this context, would mean some kind of statistics i.e. a survey of many such images and showing a correlation between the 2D Fourier transform and the malignancy of the lesion
Hey @MikeMiller I'm trying to prove some lemma for the discussion of Chern classes. If I have some symmetric polynomial, I want to establish that it is invariant (under the Adjoint action on its arguments) iff it is invariant under the adjoint action on its arguments, where Adjoint and adjoint are the representations of $GL(n,\Bbb C)$, $gl(n,\Bbb C)$ respectively.
Proving that Ad-invariance implies ad-invariance is okay. But the converse... I'm a bit stuck.
I wanted to do something like $Ad(A)(X)=(\exp\circ ad\circ \log A)(X)$ but it's not working out.
Establishing that equality is not working out---but I'm also not sure if it's the right way to go.
17:46
Hello.
$:)$
hi
@Mahmoud I forget, how did that quiz turn out?
What quiz ?
Was that you? Perhaps I'm getting myself mixed up
the math/logic one?
Yes Of course It's me, I told you that I got $\frac{13.5}{20}$ ... @Semiclassical
17:50
Now I'm having Set Theory.
And we're formalizing the Idea of a function
Hi everyone, is there an easy way to solve 2a = z - 1/z for complex z?
@BalarkaSen Tell Soham I said happy birthday
Also, uh, happy early/belated birthday to you
@Semiclassical Can you please graph $f(x,y)=x+y$ for me ?
@JamesH Multiply both sides by z, use quadratic equ?
@Akiva I haven't seen Soham in a while.
17:57
Yes of course, thanks!
But will do if I do.
He's probably busy with his exams.
Hi semiclassic I missed your reply early, I mean the integral transform mentions "Each is specified by a choice of the function K of two variables," which exactly what pixel is define f(x,y) , and about statistic, yes, we can't exactly detect charateristic of object in image perfectly. The idea is just extraction of the statistics properties of many images, so when an image is tested , it's actually just compared to the previous images mentioned.
I want to know if there will be any way to study functions like this ^$$\tan \left(x^{\sin \left(x\right)}+y^{\cos \left(y\right)}\right)=0$$
18:09
@Semiclassical, if only there's someone like you in such group like this, everything will be easier, sadly we don't have one.
18:24
eh, anyone with a background in fourier stuff would be enough
19:20
Here's a terminology question I should know. What's the phrase "cohomology class of a hyperplane" supposed to mean? (I know what a cohomology class is, and what a hyperplane is, but I don't get what the intended mapping is)
19:48
The cohomology is mathematic mapping (but as I said i'm bad engineer coz weak in mathematical background), but the hyperplane is something about class separation (clustering) when I 'classify the lesion on skin'
so called support vector machine, the term in computer science,
oh, uh
that wasn't directed at you :/
the idea is finding optimal separator (the vector) between 2 dataset
though I did wonder if there was a machine learning context to this
Oh, LOL... I thought you're talking to me haha
lol, no worries
machine learning is something I know a bit about, but not in any depth
which is too bad. seems neat.
19:53
But the difference is, you could probably explain most of the things on the field
maybe
I have heard that a lot of physicists/applied math people are able to make the transition to machine learning fairly easily. most of it's just a matter of translating the language rather than learning new concepts.
yeah, like my father
but the picture above is somehow wrong, he refers the vector as the dataset
but it could also, as dataset is just set of elements
maybe
.
is algebraic topology necessary for the study of algebraic geometry?
Pretty essential, as I understand it
ok, and in what order should i study the following
20:01
Though given that I don't have a durn clue what a scheme is, I probably shouldn't weigh in
Commutative Algebra, Topology, Algebraic Topology, Algebraic Geometry
my knowledge of algebraic geometry is decidedly shallow
so I really can't advise.
I am trying to understand this equation with standard notation. This is an ascii version of the equation
4⋅∑k=1106(−1)k+12k−1=4⋅(1−1/3+1/5−1/7+1/9−1/11…).
my math isn't up to par, but i'd like to figure out how to calculate with the sigma notation
the examples ive found online are leaving me with more questions than answers.... how would you write it by hand?
20:19
first off, you can read/express math symbolically here. use the chatjax link in the room description
once you do that, you should see this symbolically: $\sum_{n=1}^\infty n^{-2}=\frac{\pi^2}{6}$
presumably, you want that last expression to be $4\left(1-\frac13+\frac15-\frac17+\cdots\right)$
and you'd write it in sigma notation as $4\sum_{k=1}^\infty \frac{(-1)^{k+1}}{2k-1}$
which is presumably what your first expression was intended as
aah ok. Yeah, I really wasn't sure where to start, this should put me on the right track. Thanks Semiclassical!
Semiclassical, one more question... How do you know that k = 1... instead of k = 1106 ?
20:42
@arctictern here?
@meow-mix imo (topology + commutative algebra) and then algebraic geometry if you are intersted in learning algebraic geometry
if you want to know algebraic topology its enough to have a little bit more than basic topology knowledge
00:00 - 21:0021:00 - 00:00

« first day (2284 days earlier)      last day (3031 days later) »