#groups-rings-fields
1 messages · Page 123 of 1
There’s a chrome extension for it
Whats wrong with having solutions for self learning? In my analysis class my teacher gave us the instructors book
Can you guys give me a chrome extension that plays subway surfer under y’all’s comments?
or I can make only certain letters bold to make some secret messages or something.
Wdym y'all can study without some obscure video essay as background noise
I've found that it's much better to not have solutions at least in the beginning, though much of my work is "prove X" at this point which makes it easy to know where I'm starting and where I'm ending.
literally all of these extenstions sound like key loggers
I will do without
I think a parrot pecking at my laptop is enough distraction for me
@delicate orchid embrace the surveillance state
I'm such a NPCpilled psyopcel
Say good bye to privacy
say goodbye to piracy
Lol
I think the book publishers are gonna stop piracy purely by making their pdf's too big
how can you read in all bolded words? I can't get past how lumpy and extremely jarring it is to read.
it's not lumpy it's so much quicker and smoother
all these neurotyps in MY channel smh
Probs forces then to slowdown while reading while maintaining a regular reading pattern
Compared to whatever they usually do ig
Hail galaxar 🫡
no, it lets me read at the speed my brain goes at, it's faster
I feel like I'm in a twlight zone episode
literally learned a different language in order to use a logographic script instead of english so I don't have to read long words you wanna call me neurotyp again?
yeah, neurotyp
Standard textbook speed?
The worst insult in the abstract algebra channel
oh and the miles davis' ones
Slow
badada ba bapppp [13 minute drum solo] bebababobo
that's the wrong color for miles davis
As soon as I see that listing I assume chat gpt
Bada bee bop bop badap bopp
Lol
no no the one that goes be baba bap baba boop
Blah blah blah.
That said, a list of blah is
1.
2.
3.
4.
Another paragraph reiterating the first one
chat GPT uses the oxford comma
so... based...
Where is the oxford comma lol
Which one? Chinese?
last paragraph
Yes, if K is a p-group, where every proper subgroup is contained in H and [K:H]=p, then we can say that K is cyclic.
To prove this, let's consider the case where K is not cyclic. Since K is a p-group, by the classification of finite abelian groups, we can write K as the direct product of cyclic groups of prime power order. Let's denote these cyclic factors as C_{p^{k_1}}, C_{p^{k_2}}, ..., C_{p^{k_r}}, where k_1 ≥ k_2 ≥ ... ≥ k_r > 0.
Now, let's consider the subgroup H of K such that every proper subgroup of K is contained in H. Since K is not cyclic, it means that at least two of the prime power factors, let's say C_{p^{k_i}} and C_{p^{k_j}}, are present in K. Without loss of generality, assume that k_i ≥ k_j.
Now, let's construct a subgroup M of K, which is isomorphic to C_{p^{k_i}}. Since C_{p^{k_i}} is a cyclic group, it has a unique subgroup of order p. Let's denote this subgroup as M. Since K is the direct product of cyclic groups, we can write K as K = M × N, where N is the remaining part of K after removing M.
Now, since H contains all proper subgroups of K, it must contain M. But this implies that the index [K:H] is greater than p because [K:H] = [M × N:H] = [M:H ∩ M] × [N:H ∩ N] = [M:H ∩ M] × [N:H]. Since H contains M, we have [M:H ∩ M] = 1, so [K:H] = [N:H]. But [N:H] > 1 because N is the remaining part of K after removing M, and it contains at least one cyclic factor C_{p^{k_j}}. This contradicts the assumption that [K:H] = p.
Therefore, our assumption that K is not cyclic leads to a contradiction. Hence, we can conclude that if K is a p-group, where every proper subgroup is contained in H and [K:H] = p, then K must be cyclic.
oh lawddd the conlangers are here
Lol
I sthis chat gpt
Yeah
obvious. Consider the subgroup graph
In general, ChatGPT's writing style does not rely heavily on listing items in an essay format. Instead, it tends to generate responses in a more conversational manner, with coherent paragraphs and sentences. However, depending on the context and prompts provided, ChatGPT may occasionally incorporate lists or bullet points to organize information or present a series of related points.
It's important to note that the specific writing style of ChatGPT can vary based on the training it has received and the patterns it has learned from the data it was trained on. While ChatGPT aims to provide informative and helpful responses, including organizing information into lists when appropriate, it is ultimately an AI model and its writing style may not perfectly align with typical human writing styles or preferences.
The mods can’t help you now
the helps can't mod you now
I wanna program a mathgpt
shut up nerd
@chilly ocean train it off of people who ask < 3 questions on this server
Surely irony defence works
That would make for a very good masters thesis
furry gpt
no chatgpt posting folks
@elder wave are you trying to kill #groups-rings-fields
I wasn't involved in this chat btw
There was some context to it so whatever for now but don't do that
We have to rise up against the mods. When they came for foundations I said nothing because idgaf…
I would throw a whole paragraph in sp in here but that's way too off topic for this chat, and also annoying to use a bot in another server for.
Algebra has words, conlang's have words
what's the actual proof btw
algebra is the study of mathematical structure, it's just grammar but for math
Induction, centre of a p group is not trivial
Basically if you can get K abelian you are done
swag
What’s it called
And K/Z(K) is a smaller p group
(P is a p-group, [P:Q] = p, and H < P implies H <= Q) implies P cyclic
If K is a p-group, where every proper subgroup is contained in H neq K, show that K is cyclic
the good language. pronounced "toki pona"
This is actually true
hm
yeah I mean I can see it's absolutely obvious by just pointing at subgroup lattices
Why?
Why can't it be a mess lol
contains at least one C_p \times C_p right
hmm
ok so that does it for the p^3 case
oh I’ve heard of that, how do you deal with the word for “simple” being the same as the one for “good”
So K/Z(K) is a smaller p group, and every proper subgroup of K/Z(K) is contained in H/Z(K) by correspondence theorem
ok then we induct on order
swag
Then you do induction
Oh okay that's nice
Also if K is abelian, K is cyclic
By finitely venerated abelian grapes
actually does it do it for p_-^{1+2}? that has exponent p^2
I was wondering how you'd use induction but using the centre makes sense
hmmmm
I guess the idea is just you want to pick smth liek canonical to reduce the size and for p-groups the centre works?
context, primarily. We should probably take this to #chill if you wanna talk more.
oh I'm not using the centre at all, I have no idea how that would go
Can’t you prove this directly?
Because if g is an element which maps to a generator of G/Q then <g> = G
nah it still works :iwon:
It actually seems to have nothing to do with being a p-group..
Hmmm, nice
It’s a much weaker condition than being a path
that reformulation only holds in the p-group case
The point is that you can’t have a single maximal nontrivial subgroup in a group.
it's like having a little node at the top and then the rest of the lattice exploding off of the one beneath
I do think the restrictions force it to be a p group, cause if it's not I think Sylow will make it so such a Q doesn't exist
In the finite case
@frigid lark that’s true
index p p-subgroup => p-group yurrr
@delicate orchid we’re not assuming Q is a p-group
But it will have to be if it satisfies the condition in the theorem
I'll just let you two get on with it 
ok no I won't actually
I've re-read this like three times now
what does this show?
why?
what?
thanks for coming to my ted talk. I will now enjoy my ham and cheese toastie
quite a witty quip
white a quitty whip
If you want, I'll send you my proof tomorrow
It's too late rn
it's fine dw I will have long since forgotten by then
But Topos' proof is league's better
ok got it now
Shouldn't this say "where p'(x) is a product of irreducibles in both R[x] and F[x]"?
@frail summit yes.
Oki, thx
If A and B are similar CSAs (there is a division algebra with A\cong M_n(D) and B\cong M_m(D)) and C is a simple central artinian algebra (not necessarily finite-dimensional), are A\otimes C and B\otimes C similar? I know this holds if C is finite-dimensional, i.e. a CSA.
NVM, D is necessarily finite-dimensional, so D\otimes C is artinian.
Is it correct to say that if K-algebras A and B are isomorphic as rings, then they're isomorphic as algebras? K is a subring of either centre and A\cong B\implies Z(A)\cong Z(B), so the K-actions should be the same, no?
is that true? let's say that you have a field K and a (strict) subfield L which happens to be abstractly isomorphic to K. Then, you can consider K as a K-algebra in two ways: it acts on itself by the identity map K --> K, and it also acts on itself via the composition K \cong L \subset K
K is isomorphic to itself as a ring but I don't think these are the same K-algebra structure
Yeah tl;dr an algebra structure on a ring is given by the ring structure together with an embedding of the field into the center of the ring, and you can have different embeddings of K into Z(A) which produce non-isomorphic algebras
What's an example of such K and L?
Something to do with transcendence bases?
you can do that yeah
Because i can't think of any number fields like this
like K = C(x1, x2, x3, ...) and L = C(x2, x3, ...)
you can even take K = C which happens to be isomorphic to a strict subfield of itself
this wont work for anything finite dimensional over Q
I think R is a Q(root(2)) algebra with root(2) -> +-root(2), and the algebra structures aren't the same (though I don't remember why lol)
how do you figure that?
there is a theorem that C is teh only algebraically closed field with characteristic 0 and cardinality of the continuum
the algebraic closure of the field C(x) also has both of those properties
so it's isomorphic to C
so you can consider the inclusions
C --> C(x) --> \bar{C(x)} \cong C
It’s easy to find ring isomorphisms which are not K-algebra isomorphisms as long as K is not Q. But if A, B are CSA’s and they are ring isomorphic then they are morita equivalent thus equivalent in the usual sense
Then the theory of morita equivalence for CSA’s tells you that there is a K-algebra isomorphism
R is a Q(sqrt(2)) algebra with all the operators you'd expect because Q(sqrt(2)) is a subfield of R. I don't know what you meant by root(2) -> +- root(2)
@glossy crag wouldn’t it be the Swiss guard bursting in?
i think they meant Q(sqrt(2)) can act on R where sqrt(2) acts by multiplication by -sqrt(2)
i'm too inept to edit Cantor's face in. why Swiss?
yeah exactly, basically two different embeddings of the field Q(root(2)) into the ring R
oh I see I got it backways
my apologies
np
@glossy crag because a cardinality is the domain of influence of a cardinal
Ahahahaha, good one, didn't click at first
Swiss guard didn't register as the papal guard and I thought your implication was Cantor is Swiss (which he isn't)
bump lol
Well in one sqrt(2) has a square root and in the other it doesn't. So that's why they're different.
I'm guessing it's supposed to say sigma(1) = i and sigma (2) = j ...?
@south patrol what is your question?
If R and S have are rings of lengths n and m (as left modules over themselves), then R\times S is of length n+m, right?
Yes
yeah you just take the left doodad x S and then the right doodad
Yeah, just checking.
Do you concur, @delicate orchid?
Ah I assume so
Thank
what do they mean by indeterminate over F?
Freely adjoined variable. I.e. x is transcendental over F.
ah ok, thanks
Hi guys! I have the following problem. Let $G$ be a compact group, $V$ a complex vector space (finite dimensional), if $f: G \times V \to V$ is a group action that is compatible with the structure of $V$ and $\rho: G \to GL(V), g \mapsto \rho_g(v) \coloneqq f(g,v)$ is continuous then f is continuous. Now this is what I've got, if $U$ is open in $V$, we can write $$f^{-1}(U) = \bigcup_{v \in V} {g \in G : \rho_g(v) \in U } \times \bigcup_{g \in G} {v \in V : \rho_g (v) \in V } $$
gabo.brg
I have problems showing that $\bigcup_{v \in V} {g \in G : \rho_g(v) \in U }$ is open in G
gabo.brg
That pre image looks wrong
It should be the set of all pairs (g, v) such that ρ_g(v) ∈ U
This set is smaller I think and need not decompose as a product
As for the solution, you'd have to explicitly make use of the topology on GL(V). Endowing V with a norm and using the operator norm might make your life easier.
help i proved a-c and i can't put shit together through d
im assuming that there exists a in R such that sigma(a) = a' such that a \neq a' and trying to derive a contradiction
can't find any
i know that any automorphism carries positive numbers onto the positive numbers and negative numbers onto the negative numbers, but it may be a permutation of the positive numbers and a permutation of the negative numbers
i'm not sure how the ordering argument in c is supposed to help with d
Think about things in ℝ which must be fixed by the automorphism
Do you know any such elements
Nice now infer that more elements are fixed
ah i suppose i can consider the cases for b = 0
And repeat till you get everything in ℝ
t. Shing-Tung Yau
t. yer dad
imagine a Hilbert space which has 2 coordinate axis. now we got a state vector in this 2 dimensional space (like spin). now it's obvious that state vector has images on every axis. consider a rotation which conforms the state vector on one of the coordinates. then there is not any image of state vector on the other coordinate axis.
is there any rotation defining this ? ( yes )
now can we assume any physical instruction to explore the rotation which means being certain about state of quantum system ?
lmao
i would have never got this in a million years
also seems to assume that you know that Q is dense in R
Yeah so what I was hinting at is that from 0 and 1 being fixed, you find that all integers, and hence all rationals are fixed. From there, you don't necessarily have to take sequences like this. Instead, consider a real number r. If it is greater than some rational q, then σ(r) > σ(q) = q. If it is less than some rational s, then σ(r) < s. A real number is uniquely determined by the set of rationals less than it (this is tautologous if your definition of the reals is using dedekind cuts), so σ(r) must be r.
Is almost the same but I'd say conceptually simpler and there's a simpler picture of this with the number line than if you took a sequence
Oh was just smth about shuffles like
Smth I read seemed to make no sense basically lol
But probably a typo
Not sure if I should email about it lol
You might want to ask this on the physics server instead
I'm not dumb, right? for non-abelian groups the fact that there exists some ga=/=ag directly implies that this isn't a left regular action, yeah? this seems nearly trivial.
it could be considered a right regular action, if you want to define it that way (you'd probably prefer to have it be a.g instead of g.a for consistency), but it's not a LRA
Nope
nope I'm not dumb or nope it's not nearly trivial?
These are two different multiplications.
Oh it is nearly trivial, but for different reasons.
g•a ≠ ga is not an issue

What
The word regular is nowhere in your screenshot lol
it was right after a question about left regular actions
What's a left regular action 
don't worry, I still can't read
a left regular action is a group action such that g.a=ga for all a,g in G
Oh lol
So the left regular action rather
||a . (b . g) = gba which isn’t (ab).g!!!! AAAAAAAAAAAAA||
It’s the action that defines the regular rep isn’t that funny hahahahaha
The first split the field in the 1930s
see, this is why I post in here, because I'm stupid
☝️
Big tips for doing math.
🤣
given what I thought the question was, I wasn't stuck, I was done, just way too quickly. My "you're missing something" alarm went off
oh damn, yea I’m def not smart enough to think of that lol
ok, also nearly trivial, but seems much more substantial in demonstration than just prima facie defying the definition like I thought it was.
Yup.
and now for one where instead it's g.a=ag^-1, which seems more substantial to say it is a group action. I've managed to show that x.(y.a)=(xy)^-1.a but that's not quite satisfying the definitions
at least not in the obvious sense
What you write is false.
hold on, lemme do it backwards
when in doubt, read it twice and try it backwards
blech. missed something or other somewhere in there
(xy).a=a(xy)^-1=ay^-1x^-1=x.(ay^-1)=x.(y.a)
much easier going backwards
dunno why, probably got hung up on an error I thought was right and did it wrong twice.
prolly just missed an inverse or something
I love that they're asking me to deduce something that I've already previously proven for all isomorphisms (that x and f(x) have the same order). I guess the last step isn't quite trivial but it seems to follow pretty much directly from the definition of isomorphism
because it's an isomorphism F, the image of A under F should have the same cardinality, especially for finite groups
yeah these are all immediate from showing it's an isomorphism
I'm realizing that they use "deduce" to really say "this trivially follows but here's a reminder anyway"
reminder: isomorphisms are still isomorphisms
FINALLY getting to Lagrange's theorem lol
"using the previous two exercises, prove Lagrange's theorem"
(this is using 18 and 19, which I haven't shown)
alright
is the intersection of any two orbits under an operation empty?
~~Intersect it with the same orbit
~~
oh hush. I meant distinct orbits. I've tried a couple things and I'm pretty sure the orbits under an action constitute a partition, but something about the definition of an orbit I'm given is weirding me out regarding the identity of the group
Again, try proving/disproving it.
oh, duh. equivalence relation, transitivity of equivalence dictates that if 1 is in every orbit, then the orbits would be non-distinct.
Yup.
man I really be asking the stupid questions today. Now I know what it feels like to have been that kid in class
lol nothing wrong with asking questions, the annoying kids are the knowledge-shamers
it's a bit hard sometimes when the text gives you "show this equivalence relation, by the way, this is called an orbit." "now show that for any arbitrary subgroup H in a finite group G, H is in bijective correspondence with each orbit under H on G." "now deduce Lagrange's theorem from these two statements" after I've known what I'm dealing with for approximately 5 minutes
That's the point, you learn by doing.
If you spend time with them – and resist the urge to look at solutions – exercises are extremely rewarding.
it is. I just prefer to have definitions in the body of the text rather than buried in the exercises
you also develop a sense of the right questions to ask, like showing that certain subsets are disjoint you'll start to think "oh, are these things just the classes of some equivalence relation"
Sorry I realised I should've taken that to a help channel my bad
no it was fine here
oh ok
yeah I was getting stuck on where to start for part c on hw1 pinned here
I'm a newbie to abstract algebra so excuse me if I'm miss some obvious things 😅
think about noncommuting elements of S3
we can have x, y with finite order and xy infinite order
in a finite group?
oh lol
yeah this works
S3? (unfamiliar with notation sorry)
is that the symmetric group
symmetric group on 3 elements
Wrong operation 
Love the idea of calling people in help channels patients, I’m gonna steal that
is a counter example good enough for that question or would I need a rigourous proof of some sort?
a counter example is a rigorous proof
Not forall = exists not,
So a concrete counterexample is more than enough
how would i go about showing associativity for this example? do i just draw a diagram of arrows and put paranthesis around it or some shit lmfao
i mean it just follows from the axioms of morphisms in C right
kinda too lazy to draw everything out
and give names to all thes emorphisms
but oh well must be done
Tbf composition is just induced by that of C right
yeah true
oh yea facts
Also lol rare case where cat theory isn't in cat theory rather than other way round lol
Lol
bro there are like 16 million morphisms in this example that i have to give names to holy shit
huh
Oh lol I meant the q you gave was cat theory
but often I see people ask basically algebra questions in the cat theory channel
oh lmfao
nah i feel like this shit way too elementary to ask in that channel
that channel intimidates me
Hungerford
this is good enough for showing existence of identity right lol
idk how to write a curly D
wait
oops it shouldn't be g
wait
now i'm confused
ignore the clearly
I'd write it differently
Well I don't think you've technically show that this is an identity
yea i haven't
this is confusing lol
i mean it's just a matter of definitions but it's so weird
#category-theory proofs tend to be like this tbh
not in this case. This one is literally just ||1_Af = g1_B => f = g|| lol
or
the other way
you know what I mean
anyway what we doing defining the arrow category
that's the identity, but you'd need to show it's the identity with a picture more like this to show it's actually the identity morphism
hm okay thanks that makes more sense
lol I was a bit sloppy in paint but that's the rough shape of it
but everything just works out because the underlying objects are proper morphisms
thank you, yeah i'll definitely revisit this example later
That's Scriabin as your pfp, right
Any recs? I've listened to some of his stuff, but not too much
It was good ofc, but left no lasting impression
Poem of Ecstasy was good
i really enjoy all of his piano sonatas
if you liked poem of ecstasy, i would highly recommend his piano concerto, fantasie in b minor, sonata number 2, or sonata number 5
his first symphony is also like poem of ecstasy
sorry this is off topic lol
You read my mind
myself i'm a RachChad
Fellow Chad?
can you feel my heart?
Take me to the land ?
SCRIABIN
BASED
"what fantasy?"
would herstein's topics in algebra be too difficult for beginner in abstract algebra?
would it be better to stick to gallian or fraleigh?
Just start reading and you'll know soon enough
shuwi in #groups-rings-fields 

@coral shale should people not read books to work out if they're at their level or not?
Hey det, hope you're well
I got a copy of Matsumura recently. I might read it a bit over the summer
i wasnt judging, dont mind me
I will never figure out the tone of emojis in this server, apparently. "
"
det no understand that emoji either >.<
me was planning that too 
i read the first 5 sections and i'm convinced it's not a cute book >.<
Also got Rudin's book on fourier in groups, so that's cool. But it's also written by Rudin
wasn't a fan of most proofs so far, but it had some cool stuff
It looks intense
it issss 
It's a bit of a pity but I'm going to have to focus on another book over the summer mostly
Probably Geck's book on algebraic groups
But hey I should have some time for maths less important to my research
konnichiwa 
I had planned to read a lot of things lol, didn't do shit
same uwu
i can only manage like a third of what i plan 
that's still a lot than what I have done till now lol
(oh should have said that i don't plan a lot >.<)
all reps are confoosing 

commie ring thy
ah
that's what chmuwu recc'ed
CA just looked even harder
but it works too well (at a basic level) it feels like
They never did click with me. I find it harder than with ordinary modules lol
yeah bcz there weren't any good CA books available back then
maybe I'll peek at Atiyah if I get stuck somewhere in CRT
For a second there I thought "what ofc they're going to use the chinese remainder theorem"
yeah what's up with Ad, ad whatever, notational hazard
not even associative lmao
idk why I'm reading it tho
idk why i read math either
anything differentiable is scary (taking differentiable manifolds next sem...)
I probably should read some K-therory or something

what is K theory
you study the letter "K"

ah thanks for the clarification
I still struggle
like in highschool it was like "oooohhh
e^(i * pi) = -1"
that feeling has changed a bit :p
how so?
K theory
to prove anything you have to develop like so much theory 
It's only amazing when you don't know what the LHS even means
is it like "ah there's another thing I need to read up, god damn it"
e^(i pi) = -1 + ai 
$\tau$
bladewood
not fully, but a lil yea
Chmonkey
chmuwu 
I’m very chmusy
He's literally a pokemon. He's a pokemon.
you become numb and no longer feel anything 
kinda like that
I still get amazed by some topological results, not so much in algebra
for example, last thing that really amazed me was Rochlin's theorem
people say you get a nice feeling when you work hard on a problem and then eventually get it
but that's just a bunch of lies 
well
I think the classifications that Coxeter–Dynkin diagrams provide is pretty amazing. I'm still in awe over that.
I am yet to reach Dynkin diagrams
Arguing about tau vs pi is the most pseud thing you can do. Talk about actual maths, not the aesthetics of people who do maths.
but the proof is 
either you get and feel nothing, or you work on it for a while and never get it and then someone gives a hint and now you feel like an idiot 
how much math have you done?
i finished undergrad math major
die a hero or stuck long enough to feel like an idiot
going to be a grad student next year 

i hope i do too, knowing i won't
I don't know about that, being a grad student is pretty fun
It's sincere 

I proved something really nice about Schur indices the other day
If a character induces irreducibly then there's a nice pincer between the schur indices
Not the most useful but oh well
If R is a local Noetherian ring that is complete wrt some ideal m, is m necessarily maximal?
if it's complete wrt m i think it should also be complete wrt m^2
hm
this doesn't sound very true
why so?
(Here I'm talking abt the m-adic completion)
the natural map R --> lim (R/m^n) an iso right?
and m^2n would be a coinitial thingy
so they have same inverse limits
yeah but how is it an iso if you only count the even powers in the inverse limit?
basically the idea is that information mod m^n is redundant when you also know it mod m^n+1
so knowing for arbitrarily large powers will give the same thing. rest is just a diagram chase proof 
easiest way to see this is that they induce same topology so same completion

Hi, I am new to abstract algebra. Can somebody help with my problem?
Suppose $$ordG = 2n+1$$ Prove that for any a in G $$a = m^2$$
I tried to use property that states that ord(a) divides ord(G) but I don't know what to do next
mathlover
what is m?
Here's how you use the fact that $\text{ord}(a) \mid \text{ord}(G)$. Consider $a^{\text{ord}(G)}$
spamakin
just show that every element is a square
Oh right I see. Nicely spotted Spam
And then we get that $a^{\text{ord}(G)}=e$ and $a^{2n+2} = a$, so $(a^{n+1})^2 = a$ ?
mathlover
perfection
Thank you
can anyone help me understand that last sentence? I don't understand what the author means by saying that F(c) is unique up to isomorphism, because as I see it if there's another such "smallest" (being contained in any other field that contains both F and c) then it would just be equal to F(c) since they'd contain each other
there can be more than one minimal field
take like a ring
It can have more than one biggest proper ideal
(this is the same reasoning more or less since the set (or category idk) of fields is partially ordered by inclusion)
Guys a question the subgroup of R* generated by 7, would be all the exponentials of 7 ?
but how? If F(c) and say K both contain F and c and are "smallest" then by the definition of smallest above shouldn't they be equal because they are subfields of one another?
OK
what does smallest mean
It means that there isn't another field between the two you're looking at right?
but fields aren't stacked on top of each other exclusively
They can overlap a bit and whatever
Yeah, this is a very strange thing to say. F(c) is unique. Not just up to isomorphism
should be, yea
The intersection of two field extensions is again a field.
here I was just trying to convey the idea that there need not be one minimal field over another
,, ==\iso
Guess i'll ignore that sentence and go on then...
generally speaking, when is "A/B + B = A" okay?
is it at least fine for free R-modules
finitely generated free modules
when they're not finitely generated there's something called the Eilenberg swindle that sorta breaks this
wait in that case
why is
this okay?
H_0 is free, but definitely not finitely generated
This is very rarely true.
For an example, take the Z-modules A = Z/4 and B = Z/2.
I think there's some splitting of SES going on here
The only case in which it is always true is when your ring is semisimple. There are lots of nice semisimple rings (e.g. lots of group algebras, and all fields) but in general this simply does not hold.
This is also very much not true even when A and B are f.g. free
In fact taking the Z-modules A = Z and B = 2Z, it obviously does not hold that Z is isomorphic Z/2 + 2Z
Yes it is, it is isomorphic to Z.
The key thing here is that A/B is projective. Then it works
In fact every Z-submodule of Z is free.
what about free f.g. modules over a unital ring
actually nvm lol, ur example alrdy does that
Yeah.
As jagr mentions, this holds in general when A/B is projective. As it happens, any free module is projective.
i see
Z-submodules of free Z-modules are free
This is a highly nontrivial fact, but you can take it for granted.
In any case the quotient A/B here is Z, which is free and hence projective
(this is usually called the structure theorem for free modules over PIDs)
ah, so the criterion is that the quotient needs to be free
ye
Projective*, but yes.
tbh when I said f.g. free I was thinking like R^n/R^m + R^m 
wait so this requires the structure theorem for free modules over PIDs??
hatcher just states it so nonchalantly
No it doesn't here, as it happens
you just need the fact that Z is a free Z-module. Remember, we only needed that the quotient is projective.
If you really want to you can look at the proofs that this holds when the quotient is projective
The key terms are "short exact sequences ending with projectives split"
Guys, could someone give me an example of a proper inverse?
can you give more context about inverse vs proper inverse?
This doesn't use the term 'proper inverse' though. What do you mean by it?
F
yo
stupid question
let R be a ring and M be a maximal ideal
consider the R-module R/M
why does this have no submodules?
If you want you can translate, "inverso propio" and you will see that it appears as follows heh
oh yeah cuz a submodule owuld have to be an ideal
right?
Uh
the proof I know uses isomorphism thrm
where you show that simple modules are isomorphic to R / M, M maximal
I think?
Yeah, and R/M is a field since M is maximal. Therefore it can only have 0 and itself as ideals
Ah
yea i just got stupid at the fact that a submodule would have to be an ideal
so now if R/M were to be a free module
it must be of rank 1 right
cuz if it has 2 basis elements for example then one must be a submodule
Rx_2..
is that correct
...Oh.
It's reflexive (given) and transitive (by construction); you just have to prove symmetry
The given condition is not transitivity lol read it again
oh wait it's slightly different
It has to be modified slightly because reflexive transitive does not imply symmetric
Eg partial orders
In fields like (Z/3Z[x])/(x^2+1), is there any way to keep or assign degrees to these polynomials in the quotient?
What do you mean by 'keep or assign degrees'?
A lot of the nice properties for polynomials over integral domains follow because of the niceness of the degrees, I'm just wondering if one can make these carry over somehow
Any polynomial in (ℤ/3ℤ)[x] is represented by a constant or degree one polynomial in the given quotient ring.
Yeah, but e.g. is x+1 irreducible in the quotient too?
I'm sorry, but I don't understand what your main question is here.
In (Z/3Z)[x] we can immediately conclude it is irreducible because it's a linear factor
it's a field, no elements are irreducible
And degrees add up
eh...
part of the definition of being irreducible is being a non-unit
Oh ye
Oke, but in general, if we take the quotient and say some element is not a unit
Prolly just me misunderstanding smth, nvm - thx tho
it's just a ring it doesn't matter if it's a quotient
you can use all the regular methods to see if an element is irreducible or not
i.e. is the ideal generated by that element maximal etc. etc.
I think what you're looking for is the theory of graded rings. In order for a quotient of a graded ring to be graded, you need a homogeneous ideal. Unfortunately (x^2 + 1) isn't one, if I'm not mistaken.
ah yeah gradedness could be a good generalisation
I believe an example of when you can have the degrees descend to a quotient is k[x,y]/(xy-x^2). I'm rusty on anything to do with graded rings, so perhaps someone can correct me.
Noticed a huge mistake, do forgive me lol
it isn't, homogeneous ideals in polynomial rings are generated by homogenous polynomials (if it's the same definition as in alg geo)
((big if))
Yeah I corrected it to a homogeneous polynomial
yur
I guess that example is a bit boring though
maybe a nicer one is k[x,y,z,w]/(xy - zw), that's kinda interesting
coordinate ring of some plane in k^4.,, sopoppy
Found the section in Dummit and Foote covering graded rings - will mark it for when I get there. Thanks
OK nerd............ Z[x, y, z]/(x^3 + y^3 - z^3) what ya think of them apples
that's like a sphere that's somebody denotated a grenade inside
or is that +z^3
irrelevantgardless the shape is similar
u just intersect it with the lattice
the algebraic nonsense will be very different
but I do not care
But wew...
there are no solutions to x^3 + y^3 = z^3 in the integers where at least one is nonzero
:)
good thing there's a minus sign
we're looking at x^3+y^3-z^3 = 0 not x^3+y^3 = z^3 you fool
You forgot to say Uno, this doesn't count
Do you mean "... where all are nonzero"?
Lmao yeah said the total opposite
I was wondering how $|a_0|=|\Pi|^n$ implies the index is $\geq n$, does this work?
Suppose $z$ has infinite order and $m=|G:\langle z\rangle|<\infty$, we want to show $x^n=z\implies m\geq n$: since $m$ is the index $x^m\in\langle z\rangle\implies x^m=z^k\implies z^m=x^{mn}=z^{nk}\implies m=nk\implies m\geq n$. Now apply this to $|E^\times|\supset|K^\times|$ and $|a_0|=|\Pi|^n$ ($|a_0|$ is a generator of the infinite cyclic $|K^\times|$).
leave_no_norm
Hello I have a problem and I want a hint. We have a commutative group G of order n>=2 and n is not prime. Let m = the greatest proper divisor of n. Let p be a prime number. If we have at least m elements from G with order p, then all elements from G{e} have order p.
try classification of finite abelian groups
I think I know how to do it
If I take H the set of elements x in G with x^p=e then |H|>m. H is closed (because of commutativity) and finite so H is subgroup of G and |H| | n, but because |H|>m then |H|=n so H=G.
Hi, I calculated the minimal polynomial of $a = \frac{1+\sqrt(5)}{2}$ and $b = (z+z^{-1})$ over $\mathbb{Q}$.
Where $ z = exp(\frac{2\pi i}{5}) $
In this context i want to show that $\mathbb{Q}(a) = \mathbb{Q}(b)$.
Does one calculate $z+z^{-1} = \frac{\sqrt(5)-1}{2}$ explicitly here and show that both fields are equal or how does one usually approach this (when working with roots of unity)?
Also if it wasnt for wolfram alpha, I wouldnt have known that $z+z^{-1} = \frac{\sqrt(5)-1}{2}$. "Should" one be comfortable calculating these in the context of abstract algebra (and later commuative algebra and whatever else might follow)? As in does this type of knowledge come in handy every now and then?
aabb
you should only be comfortable calculating these things if your exam tests you on them
otherwise just use a calculator for these "trivial" calculations
but that z + z^-1 can be done using the cos + isin representation of your e^ and some trig identities
but yeah since you've shown that 2b = 1 + sqrt 5 you can reduce Q(b) = Q(1 + sqrt5) = Q(sqrt 5) and similarly with Q(a)
which shows that they are equal
(you can cheat a lil, the extension Q(zeta_p)/Q is galois with group (Z/pZ)* which is cyclic, so there is a unique quadratic extension which can be shown to be Q(g) where g^2 = legendre(-1, p) * p, and noticing that for odd primes, L=Q(zeta + 1/zeta) is the unique extension with [Q(zeta_p) : L] = 2)
I think we will do this in my next alg nt lecture
we did cyclotomic fields and legendre symbol last time
primes in extensions are still confooosing 
@rustic crown did you know that graded means graduiert in german
nope
weird word
det no know any german 
you know zwischenkörper
Yeah im not sure If my tutor allows this 
yee but if you wanan do it by hand, notice that z satisfies
1+z+z^2+z^3+z^4 = 0
divide by z^2 and write it as a poly in z+1/z, which gives you a quadratic in that and solve it uwu
what does "semigroup in ..." mean? also what exactly is a semigroup? I thought it was an associative magma (wikipedia), but my prof is saying ".... is a semigroup in ... with neutral element 0" so I'm confoosed, shouldn't it be a monoid then?
all monoids are semigroups
Why not
idk
You don't have to specify the most restrictive thing something is.
also what about the "in" part
You've omitted context so I have no idea.
I presume that your prof meant it was a subsemigroup.
(oh i was thinkign semi-group object in a category
)
That might also explain why they didn't call it a submonoid, as typically submonoids are defined as sharing the same unit element as the thing they sit in.
(but that makes more sense)
let $\nu : A \setminus \set{0} \to \bQ^r$ be a valuation. then $S(A, \nu) = \set{v(a) | a \in A\setminus\set{0}}$ is a semigroup in $\bQ^r$ with neutral element $0$, the so called valuation semigroup
I think this makes the most sense
oh lemme fix my texit
comic sans
fixed
COMIC NEUE
comic sans
I think he calls it semi group because that sounds cooler than monoid in german
bewertungsmonoidgruppe
bewertungshalbgruppe
ok thanks boytijejie
ermmmmmmmmm evaluates ur ring
oh no illum fell asleep on the keyboard 🙀
Thanks, also reasonable answer in general. I'll ask the assistant. Also since our courses probably looked similar (as in standard german into to algebra class), did you just grind out past papers or are there other nice resources with exam esque questions? (everyone can answer)
@formal ermine, meant as reply to your answer
I did the assigned homework
but for practice questions you can just take any algebra textbook and do the exercises in it
like d&f or artin
ah so it would look like this:
silly question but is there any way to diagram a product for a family of three objects A1 A2 A3 in a category C?
What do you mean by "to diagram"?
that's my naive thought lol
like what kryo just posted
ye that's kinda what i ended up w/ too
shit seems like a topological probelm
idk
Sure, you can draw a diagram for the product of any number of factors
you'd probably end up with different unique maps for (X x Y) x Z and X x (Y x Z) but they're probably naturally isomorphic in some sense (lol it's been a long time since I've thought about cat theory so take it with a grain of salt)
Yes, they differ by the unique natural isomorphism making the product associative
And both of those are naturally isomorphic to the usual way of defining the product of three things
no i'm not
Hm I'm feeling a little silly right now. I have a Lie group G whose center is S^1. The claim is that in any finite dimensional irrep of G, the center of G must be diagonalizable since it is compact and abelian. I'm not sure if this means S^1 is simultaneously diagonalizable, or just that the image of every element is diagonalizable. The latter is clear since every element has finite order, but the former seems more relevant and I'm not sure how to extend simultaneous diagonalization to all of S^1
i'm struggling to see exactly how F and F' being free on X and X' and the cardinalities of X and X' being equivalent lead to the existence of a bijection from X to X', could i just take i: X --> X' to be f and fbar: X' --> X (since C is a concrete category and we can consider X' and X to be functions on the underlying sets; don't really know if this works) (to show injectivity)
in this definition
or am i just hella overcomplicating things again lmfao
The cardinalities being equal is equivalent to there being a bijection between the sets.
This isn't an algebraic fact, this is just a set theory fact.
btw as an aside, why are these types of questions not asked in the category channel
I'd say they're on topic in either channel.
So since S^1 is in the center, multiplication by an element in S^1 defines a morphism of representations of G. So by Schurs lemma they are all just scalar multiples of the identity.
Also not every element in S^1 has finite order.
Ok yes I feel extraordinarily silly today. Thanks for your help
waltering
Funnily enough, it didn't matter that it was compact
oh yea oops
forgot about that thank you
holy shit right that’s literally the definition fuck
It's okay to be silly occasionally 
honk
Too late for you 
Duck duck duck goose!!
Pretty sure bijections are the definition of equal cardinality
yea i remembered tha tlater
how is tau inverse onto F bar? i really don't see it
nvm i'm stupid
wait huhhh
wait how is it onto Fbar never mind
what
the image of F under tau is F since it leaves it fixed
so how can the image of F be all of Fbar
nvm
hello SARGE!!!
i am in attendance
wtf am I readddinggg
oh finally some GOOD math it looks like
not to you i meant to what ive been going through lately
Sloth King Daminark
are cosets of open groups open dami
Yeah
oh yeah duh
Multiplication by g is a homeomorphism
multiplication by g is continou
yes
translation is continuous Why do i even bother sending messages if imbgonna get sniped
I'm reading these real fast so I might ask stupid things
Sloth King Daminark
its pretty quick iirc
that clopenussy 
Since if you contain any neighborhood U of the identity, you push U around by g in H and get that H is union of opens
And then G splits into cosets of H, each is open since L_g is homeomorphism, so G\H is a union of all the nontrivial cosets and is thus open, so H is closed
(It is not true in general that closed subgroups are open ofc)
something like H^2 = H and H bar is stuck between H and H^2 should do it
Or what I said
or what you said
this is so different to how I'm used to thinking 
but I follow completely yeah
So yeah if v_i is stabilized by open subgroup H_i
Then v_1 + v_2 is stabilized by the intersection, which contains open neighborhood of 1, so it's open
I have absolutely no idea why that last implication holds
What I said here
If H contains U contains 1, then for any g in H, H contains gU contains 1
and gU is open since L_g is homeo
ah ok
yeah so I get why these vectors are called smooth I think
it's like their "preimage under the group action" is open kinda deal
Functions on p-adic groups are considered "smooth" if they are locally constant of compact support
schwartz bruhat time
right makes sense
compact support so you don't get any nonsense "towards infinity" I presume
Sloth King Daminark
I have the notes up u don't need to latex lol
this bit I understand completely because it's just rep theory lol
ya lmao
I'm also doing it to help me think, but I guess I can just type badly 😛
this i can follow but i may not be p-adic group pilled enough specifically to get some of the later stuff
So dual of smooth isn't smooth, I'll let some grad student find an example
LOL
hmmmm
I mean it's prob like
Remember that the algebraic dual space of some infinite-dim vector space can be fucked
STrictly larger cardinality blah blah
yeah I know
it would have to be infinite dim
but there is no such thing as an infinite dimensional vector space so we're all out of luck
Tru
wow. talk over
p-adic groups are now d o n e
Actually the only irreps of GL(2,Qp) are characters in finite dimension
So as far as Wew is concerned automorphic forms are just ez
the trick is I am simply not concerned
anyway lemme think about why the smooth part is fixed
if we pick a g outside of the isotropy group and consider vg
hmmm
It's prob the general shtick that stabilizer of gv is gStab(v)g^{-1}
yeah
Conjugate of open is open gg
is preservation under surjective morphisms some standard rep theory thing im just forgetting rn
Well not just continuous but a homeomorphism, thus an open map
wowie kazowie!
Hmm let's see
it's almost like I already knew it was invertible DAMI
anyway so we restrict to that doodad and get the contragredient reppo
Sloth King Daminark
I believe the remark cause subquotients and direct sums are obvious I think
OH
So yeah surjective morphisms make stabilizers bigger
ok this is very believable
So if they were already open...
yur can't make em smaller
So yeah subrep because... it's got the same damn stabilizer
And that handles subquotients
this holds just for a general group action which is why I believed it immediately
well, T has to be a G-set morphism
but that's irrelevantgardless
And if (v,w) is in the direct sum, its stabilizer should just be Stab(v) cap Stab(w)
yeah exactly
just intersection of stabilizers and win
Which is still open
uhhhh but what if infintie direct sum ⁉️⁉️⁉️⁉️⁉️⁉️⁉️⁉️
those dont exist actually
yeah I know it still holds cause in an infinite direct sum only finitely many of the dudes are non-zero or WHATEVER
all vector spaces are fd all direct sums are finite its true
so it corresponds to a finite intersection I GUESS
Well if you take Hilbert space direct sum now you might have to actually think 😛
direct integral
no stop
anyway I'm getting distracted
Tiem 4 parabolic induction
let us peep back at the notes
time to INDUCT
So yeah condition 2 is just saying K_f becomes the stabilizer of f in this rep which we want to be open so that we get something smooth
sensible
And we should prob get that one theorem
you've been hit by, you've been struck by
a smooth inducer
Well idk about "agrees with tau" since H is acting on something different now?
mans remains unbothered
But that one thing about restriction and induction being adjoints
Whose name I'm forgetting for some ungodly reason
frobenius reciprocity
frobby recopo
guessing it still works here
as it should
if ur induction functor isn't left adjoint to ur restrction functor u got ur definitions wrong
Sloth King Daminark
I'm soyjaking rn
Sloth King Daminark
yeah I thought this held automatically dude to the gStab(woiegfsd;if)g^{-1} dealio
I suppose not cause we don't automatically have that all of V_tau is smooth
Sloth King Daminark
ne bovva
anyway I shouldn't be too surprised there's parabolic groups here
but it's something I already know about which is pog
Compactly induced geez
does the existence of the unipotent radical follow from the quotient being complete?
I'm like thinking you quotient everything again by the (???) of all the subgroups which plays nice because they're closed
and then consider some sequence and use the correspodence theorem? dunno
just spitballing
Probably some algebraic group shit?
Complete might mean complete variety here
Rather than complete metric space
what a bunch of BALONEY
This is different from what I'm used to thinking lol
oh lord there's an integral
Borel subgroup is maximal Zariski closed and connected solvable subgroup
And parabolic means contains Borel
I've literally never cared about it beyond it being the upper triangulars in GL_n
Pretty much
and parabolic to me is a subgroup of a reflection group that leaves (some set of or) a hyper-plane completely fixed
I just think of parabolic as shit like
but that's obviously equivalent to containg borel
Sloth King Daminark
yus
this would be the parabolic group fixing the uhhh
not sure
xy-plane? possible
Setwise stabilizes the subspace/incomplete flag <e_1,e_2>
yussss
that's the one I think
calling a subspace an incomplete flag...
that's gonna be a 24h mute
I'll just black box Levi decomposition
I'm gonna blackbox parabolic induction a bit 
actually I think I'm just gonna jump off here cause I got NO CLUE now
Get back here @delicate orchid
And don't think I didn't notice your disappearance @ebon gyro
you bring the haar measure in I :gottago:
So the point of slide 7 is just that, the stuff that happens on slide 6 for algebraic groups are still gucci when we pass to F-rational points, and thankfully G/P is compact so smooth = compact
I've never done abstract algebra but I have a question about the graph of this style
Or
I feel like abstract algebra is like black magic where you point out with arrows and weird symbols everywhere
the top is a definition lol
Like why do you do this kind of graphs?



