#help-23
1 messages ¡ Page 413 of 1
Well, z - z' = 0 surely works
The other unfortunately work only if the number is non 0
But apart from that they're all the same thing yeah
x^2+y^2=/0
?
sqaure of real number is 0 only when both are x=y=0
so exclude the origin my statments are correct?
I don't get what you want to achieve
@copper wraith Has your question been resolved?
Closed due to timeout
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
⢠Ask your math question in a clear, concise manner.
⢠Show your work, and if possible, explain where you are stuck.
⢠Do not immediately ping people or roles. After 15 minutes, feel free to ping <@&286206848099549185> once.
⢠Type the command .close to free the channel when you're done.
⢠Be polite and have a nice day!
Read #âhow-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
Hello, I need some help with matrix algebra. Can anyone help me?
I completed a set of problems, but the thing is the textbook has no answer key, so typically I run it through GPT or something which is fairly accurate for most times, but I believe its making a few errors.
Post away
So here are the problems I completed for Problem Set 2.3 in Gilbert Strang's MIT 5th edition Linear Algebra Textbook. Posting the images of the problems for the problem set. Please note that I did not complete the questions in a sequential manner and did not do all of the problems in the pages, just did some of them.
Also if you have any tips of general improvement for my work please let me know.
@void cosmos Has your question been resolved?
Please post images (such as PNGs or JPGs) of the question rather than other filetypes such as PDFs which have to be downloaded. Non-image downloads can potentially contain viruses or other security risks.
Hey Benze, I just got Chat to recheck it again and it is confident on most of then, so I will just then request for you and others to check on the remaining ones left that chat isn't fully sure about.
Sure, No problemđ
So if you could look at these questions specifically: (Chat isn't too sure about these ones)
3)
9(a), 9(b)
18)
19 except for the extra âfind another non-diagonal matrix whose square is
đź
Iâ part, which Iâm not fully sure you answered clearly
21 still looks incomplete to me unless Iâm missing writing that is too faint to read
Ty.
Looking at the 18th one
it could either be orthogonal
or it could be idempotent
otherwise they would not ask such a high power
For question 3 I recognize I made a silly my dumbass accidentally made it into a Lower Triangle instead of upper lol.
(Just a second I'm doin these rn)
Thanks and then compare it to the work I have in my PDF please thanks.
(.....pdf i dont wanna open i dont wanna get virused or smth-)
I'll just post images then lol no virus I've been a longtime member of this server. Almost 6 years lol.
it's not about you its the fact i dont do it anyway
Here is my work for the questions, I haven't attempted Q21 yet, so no needa check that one as I'll do that later so if you can just contrast my work to the questions for 3, 9, 18 and 19 that'd be great.
Hmmm, but because they said U doesn't that mean upper, so shouldn't I have it in the upper format?
Nah i think this is fine
They didn't explicitly mention it though in the question, so I didn't think too much of it,
I just plugged and chugged till i got it too didn't put any thought in it
Yah I think thats fine doesn't really matter honestly. Alrighty keep going
Yeah with these things they tend to get a bit monotonous, so its like just cranking away lol. I'd honestly probs double check my work, but its almost 4 am for me and I am so dead tired lol.
elimination matrices is not exactly my thing T-T
So I'll leave the ninth one to you too
I know how to do it
but I ain't fully sure about it
No worries, I appreciate the honesty.
9 looks good to me double checked it and everything works out.
n=100 in there ashte
Yep I got that.
the rest of 18 is basic multiplication
Can you check EF and FE for me pls real quick.
sure
EF is
1 0 0
a 1 0
b c 1
okay
19 is a bit weird
it could be a element switching places type
one sec
Hmm I checked PQ and QP those look good.
For M^(2) = I, I just utilized a permutation matrix and so multiplying a row exchange matrix by itslef just undoes previous operation, so we get the identity matrix.
P removes Q and Q removes P is the main idea of this thing
So that means you can either go one step forward in the pattern or one step backward
tht should be it
Hmmmm was the way I did it fine would you be able to see?
Sure if you've come to the same conclusion dont overthink it's fine
i didnt bother i saw two matrices and decided one does things and the other reverses it
np
me too
Thank you. Can I add you in case I have the occassional question or something?
I'm in highschool
you prolly know more
I won't bombard or anything, just for the occassional question.
Oh lol, I'm in Uni.
well strang is a book they teach us to some extent
Ohhhh shoot my roots go back to India, but I have been raised in Canada for majority of my life.
I've heard about JEE.
Really tough exam and I have deep respect for anyone who attempts it.
we're going off topic
.close()
(we'll meet again chill)
(dont hv to add me for tht)
Closed by @void cosmos
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
⢠Ask your math question in a clear, concise manner.
⢠Show your work, and if possible, explain where you are stuck.
⢠Do not immediately ping people or roles. After 15 minutes, feel free to ping <@&286206848099549185> once.
⢠Type the command .close to free the channel when you're done.
⢠Be polite and have a nice day!
Read #âhow-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
@midnight steeple Has your question been resolved?
Closed due to timeout
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
⢠Ask your math question in a clear, concise manner.
⢠Show your work, and if possible, explain where you are stuck.
⢠Do not immediately ping people or roles. After 15 minutes, feel free to ping <@&286206848099549185> once.
⢠Type the command .close to free the channel when you're done.
⢠Be polite and have a nice day!
Read #âhow-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
x^2 - 4 |x| + 3 < 0
My answer came x belongs to (- 3, -1) U (1, 3)
But the workbook answer says only (1, 3)
Is my process alright?
do I also need to take x^2 = |x|^2 , I thought it was unnecessary
yeah the workbook had my answer as option D but the answer showed B -> (1, 3)
alright
,w solve x^2 - 4|x| + 3 < 0
thanks for verifying
what
oh yueah
that is same as (-3, -1) U (1, 3)
thanks for verifying again
.close
Closed by @edgy latch
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
⢠Ask your math question in a clear, concise manner.
⢠Show your work, and if possible, explain where you are stuck.
⢠Do not immediately ping people or roles. After 15 minutes, feel free to ping <@&286206848099549185> once.
⢠Type the command .close to free the channel when you're done.
⢠Be polite and have a nice day!
Read #âhow-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
Tom plays a game, he is fighting the boss. Initially the boss has infinite hp, Tom has 100 hp. If Tom hits the boss once he gains 30 hp from the boss and if the boss hits him, he looses 10 hp. There is 50% chance of landing hit for both. If Tom's hp becomes 0 game over. Find probability that Tom looses
under these conditions isnt the probability just 1? because there's no way for him to win...
the boss has infinite hp
maybe
sam beam
waterrbeam
beam
@smoky sandal Has your question been resolved?
!15m
Please only use the <@&286206848099549185> ping once if your question has not been answered for 15 minutes. Please do not ping or DM individual users about your question.
That's first of all
Second of all
Someone asnwered your question and you said maybe
Then sent a meme gif for some reason
And that's very weird
If you think the question wasn't correct or you didn't understand the answer then that's fine but it feels very weird to ping for this
its been 20 mns
that was because of
???
Is this NOT your question?
yes that happens everytime i try to to ping helpers
Can you clarify your question, please?
what is not clear ?
What your question is? The pinned question has already been answered by sam.
So what do you need help with?
you can have infinite moves, so idk if its 1
Okay well let's put it this way
They asked for the probability of tom losing, correct?
But since there are only 2 options, winning and losing, we can instead say that the chance for tom losing is 100% - the chance of winning
So if the boss has infinite HP, what are the odds that tom will eventually get the boss to reach zero health?
but the number of moves is also infinite
Well, infinity is not a number so you can't have something that goes on forever
What you could have is an indefinite repition where tom always gains HP faster than the boss can damage him
But you can't reach an infinite number of moves
oh but that does not mean game over, the game can go on forever?
The game can go forever
But it's probability is 0
Your question is interesting and requires a but of combinatorics
hm.. why
Do you know how to do basic combinatorics
I'll try to bring you there
a bit
Okay good
Do you know about binomial coefficients
(Lemme chekc the english name)
yes
Ok
Good
So we're gonna first look at a simpler problem
If the boss hits and if they hit the boss, let's first assum that the hp gained and dealth are equal
ok
probability that tom loses is 1 minus the probability that he wins.
boss has infinite health so can never die, so tom can never ever win in the first place, so the probability he wins is 0
1 - 0 is 1. the probability that tom loses is 1.
The problem is thatwe don't know the probability that neither wins
0
what constitutes a draw? there's no draw state
sorry i gtg brb
Draw: the game co tinues indefinitely
probably negligible 
It's pretty hard to prove
is it even worth the effort 
then happy helping
Im going to prepare some drawings for when they come back
The way I'd explain it is just through coin flips
Like the chance of 2 heads in a row is 1/2^2
And 5 in a row is 1/2^5
And it gets smaller
Until it eventually reaches zero at infinity
Which means that it could never happen
@smoky sandal here are some intermediate questions, might not be perfect so ping me if you don't understand, need help or have doubts.
Ig question 6) is to look at @torpid osprey's argument and adapt it
- is tricky so maybe we'll have to talk about it
Hey hii everyone, honestly I am no maths freak but my girlfriend left me so I felt lonely so I just joined discord
#discussion is the best channel to talk about casual stuff
Alternatively #serious-discussion or #chill
@smoky sandal Has your question been resolved?
Closed due to timeout
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
⢠Ask your math question in a clear, concise manner.
⢠Show your work, and if possible, explain where you are stuck.
⢠Do not immediately ping people or roles. After 15 minutes, feel free to ping <@&286206848099549185> once.
⢠Type the command .close to free the channel when you're done.
⢠Be polite and have a nice day!
Read #âhow-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
For question b I used s=ut+1/2*at^2 but i got the wrong answer because I did -9.81 and the correct answer didn't use a negative. So my question is why is the gravitonal pull not negative here?
maybe because they oriented the positive vertical axis downwards
@lavish magnet Has your question been resolved?
<@&268886789983436800>
.close
Closed by @lavish magnet
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
⢠Ask your math question in a clear, concise manner.
⢠Show your work, and if possible, explain where you are stuck.
⢠Do not immediately ping people or roles. After 15 minutes, feel free to ping <@&286206848099549185> once.
⢠Type the command .close to free the channel when you're done.
⢠Be polite and have a nice day!
Read #âhow-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
That's odd. What happened?
yeah
in the future you can type .reopen to reopen the channel to avoid that happening (you closed it beforehand)
fahhhh
all g! you can click on the forwarded message to bring up whatever you said last
Sounds good.
@blazing swallow if possible can you forward other txts too
or @amber glade go to help 25 and type .reopen to chk if it works
ok W
.close
Closed by @blazing swallow
Use .reopen if this was a mistake.
maso7
Send your question here to claim the channel.
Remember:
⢠Ask your math question in a clear, concise manner.
⢠Show your work, and if possible, explain where you are stuck.
⢠Do not immediately ping people or roles. After 15 minutes, feel free to ping <@&286206848099549185> once.
⢠Type the command .close to free the channel when you're done.
⢠Be polite and have a nice day!
Read #âhow-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
Renato
do I need to apply the arquimidean property or the axiom of the supremum?
I am not sure how to proceed with 11a)
In particular, the idea is if the sequence converges to l, then eventually r should lie outside the epsilon-interval you set around l.
|r-l| in this case is the distance between l and r. Try to use this as Green suggested.
I dont get it
an -> l <=> n >= n0 => |an - l| < epsilon
what does l and r has to do with this
All I'll say is that r-l > 0 (and this should ring a bell from the definition)
what?
epsilon > 0 and r - l > 0
Indeed.
The definition tells you something about any choice of epsilon > 0.
the sequence is eventually contained in any epsilon neighborhood of l
r is some distance from l
youâre on a?
ye
in the definition of the limit this holds for every epsilon neighborhood
the distance from r to l would be a smart choice here
what is that distance?
epsi
0
that would be if r = l
ok
but
what is the distance
if you had to write an expression for it
like distance = âŚ
who knows at this point
whatâs the distance between 3 and 4?
r - l
yes
|3-4
so you know that the sequence is eventually contained in an epsilon interval with radius r - l around l
can you picture this on a number line?
place l anywhere
then r somewhere to the right of it
draw an arrow above the number line from where l is to where r is then draw an arrow in the opposite direction of the same length, starting at l
that will create your interval
then just imagine a bunch of points from the sequence being trapped inside there
?
how the hell are we supposed to know if he understands every textbook example and prior exercise
no one works a full time job helping renato
we just pop in on our own time
this is more a question to renato
"TO OTHER HELPERS:"
@spiral saddle remember i asked you to not open a channel without having at least one concrete idea of solving the problem?
can u make a drowing
what wasnât clear here?
my idea was using the definition of convergence of a sequence aswell as the arquimidean property or the axiom of the supremum
how would the archimedean property be used here?
those are all very vague ideas
the closest idea is use definition of convergence but thats still very vague
and the limit of a sequence doesnât have to be the supremum of the sequence
this sequence isnât necessarily monotone increasing
you should first ask for ideas from the professor or ppl who already passed the class
either use the archimidean property to find a lower bound for the epsilon
if I would know a concrete idea on how to solve everything then I wouldnt be needing help, just a will to execute it
it takes several concrete ideas to solve this. i simply need you to have at least one coming in
and i dont believe you asked the professor or senior classmates for ideas before coming here
you probably dont have the context of a very long convo i had with ren the other day
i donât
can I get some help or not?
2l - r?
yes thatâs just the point at a distance r - l from l on the left
it doesnât really matter here
you donât have to label it
subtract r - l from l and you get 2l - r
which translates to going a distance r - l to the left of l
i didnât draw points in the sequence outside of this interval but there can be
itâs just that there would only be finitely may of them
l - (r - l) = l - r + l = 2l - r
I see
because the sequence has to eventually be contained in this interval
why are you showing me this
a visual aid to get a better intuition for the problem
it should be clear from this image why a is true
do you see how a_n < r for all of the n pictured here
yes
now you just need to write it with symbols
how would we describe all of the sequence members being in this interval
what can we say?
(eventually)
before we go further, you understand what i said about it being possible that finitely many of the sequence members arenât in that interval right
no
ok then you shouldâve asked
|an - l| > |r - l|
i was trying to describe all the sequence numbers in the interval
no thatâs not quite right but we will get there
this is more important for now
but maybe its
2l - r < x < r
how
in the sequential limit definition there is something about "there exists N such that for all n > N, âŚ"
the finitely many indices before N are irrelevant then
why are you saying finite
we donât care if a_1, a_2, âŚ, a_N donât fit in that epsilon neighborhood, so long as we can find such an N so that for all of the indices afterwards, the sequence members are in that neighborhood
these first N makes up a finite list
of sequence members
ok, yeah because n >= N
like imagine i had the sequence:
1, 0, 0, 0, 0, 0, âŚ
clearly this converges to 0
we donât care if the first sequence member doesnât fit in every epsilon neighborhood around 0
and it doesnât have to be the first
we can have something like
ok
1, 0, 0, 0, 0.5, 0, 0, 0 etc
and now the 0.5 might be a problem for some epsilon intervals
but we can just take the points after it
which is an infinite list
yeah
so here we can take the last point that isnât in this interval and just throw that first part of the sequence out
which i didnât draw
but imagine there are some other points in the sequence not in the interval
there can only be finitely many of them
because we have to cluster around l
so the sequence essentially shrinks around l
we canât stay far away infinitely often
how do we find a suitable N?
i mean I am still stuck on how you grasped that a) is true
recall that for any epsilon > 0, we can find such an N so that blah blah blah
so if we come up with some epsilon > 0 then by definition we are guaranteed the existence of some N, whatever it is
we said r - l > 0
so by definition there is some N so thatâŚ
you tell me
hmm
not sure what youâre asking exactly
like how do i have the intuition that this is true?
iâve done tons and tons of these problems and familiarized myself with it
a lot of people just see the definitions and donât concretely understand what itâs saying
if you understand it conceptually it should be no issue
i know that a_n has to get "closer" to l than any other number which means itâll eventually have to be below r since l is below r
the definition of convergence of a sequence used that forall n in N there exists some n0 in N such that n >= n0
the sequence is clustering around l
no?
thatâs just an obviously true statement, take
n_0 = n
there is some N so that for all n > N,
|a_n - l| < epsilon
ever seen this
yes
yes but
how to connect it with r < l
we already said which epsilon we are interested in
^
r - l is our epsilon??
you have a solid grasp of the def
đđť
is it that easy for u?
this?
yes
youâll get comfortable with it eventually
itâs like tying your shoes
say i have a sequence bn that converges to l when n -> inf, then here the left side endpoint will be what? and right side endpoint?
well in general itâs l - epsilon and l + epsilon
if you want an interval around l with radius epsilon
here i had a specific choice of epsilon in mind
because i wanted a_n < r
which translates to r being the right endpoint of one of these epsilon intervals
you make it seem easy
you should get comfortable with epsilon intervals
this is the best way to think of convergence
iâve spent more time on it than you have
my fault
đ
bae?
alr I appreciate the help, though I cant say I am confident about this stuff
try the next one
this is my first exercise using the definition of convergence of a sequence
^read some of these again
where i explained my thought process
lim an = l <=> |an - l| < l - r
there are some unstated conditions there
jajaja
but yes l - r is a good choice of epsilon
sorry, how can I be more formal about it?
do you mind helping me with the wording
how about i word the first one for you
since a_n â> l and r > l <â> r - l > 0, there exists n_0 in N such that
|a_n - l| < r - l <â> l - r < a_n - l < r - l <â> 2l - r < a_n < r whenever n > n_0
ok maybe latex would be better
but i think you can read this well enough
keep it simple
if you donât like the whenever n > n_0 afterwards you can just replace it with for all n > n_0 before the inequality equivalenceâs
sinice an -> l and l > r <=> l - r > 0, then there exists some n0 in N such that
|an - l| < l - r <=> r - l < an - l < l - r <=> r < an < 2l - r
perfect
what
but just say
whenever n > n_0
at the end
or say for all n > n_0 after the such that
n >= n0?
but this proof is complete?
yes why wouldnât it be
weâve shown that there is some n_0 such that for all n >= n_0 the inequality holds
we never found one such n0
we did
we guaranteed the existence of it here
we donât need a formula for it
just need to show such an n_0 exists
we know it exists because a_n â> l and r - l > 0
or in the second case l - r > 0
and for any positive distance such an n_0 exists
which comes directly from the definition of convergence
oh right
I see
what about c)?
thanks
well letâs think about this for a second, if r >= l then r > l or r = l
we already handled the case when r > l yes?
what
what was confusing
there are two cases here
no because now we have an <= r
r > l or r = l
in the other exercise we had
an < r
ok but if a_n < r then a_n <= r
so the r > l case has been resolved
now what if r = l
is is true that if r = l then there is some n_0 such that a_n <= r for all n >= n_0
i.e. is a_n eventually bounded above by l
this just says a_n <= l for all n >= n_0
as long as it is bounded above by L then an <= r forall n >= n0
is this true if a_n â> l
sure because l = r
^
how do you know an is bounded above by l
yes
well a_n <= l means a_n is bounded above by l
and we had r = l here
similarly an -> r
dude, lim an = L
yes
that by itself means that an -> l
an -> l <=> |an - r| < epsilon
but epsilon needs to be > 0
you said an is bounded above by L
if l = r then you canât make l- r or r - l epsilon
no, i said thatâs what the question is asking
is it true that a_n is bounded above by l if a_n â> l
eventually
we need to use n >= n0 somehow and find a suitable n0 maybe?
well try thinking of it visually
uff
if i have a sequence of points converging to l, do they have to eventually cluster from below l
or can they reside above l as well as they shrink around l
they can reside above l aswell as they shrink around l
because L is not an upper bound for every number in the sequence
this converges to 0
but 0 is not an upper bound for the sequence
we are entirely converging from above here
1/1 = 1
not sure what you mean
thatâs the first member of the sequence
sure
1, 1/2, 1/3, 1/4, âŚ
what I mean is that the sequence goes
- 0.5, 0.3, 0.25
is 1 an upper bound?
start it off by saying if r >= l then r > l or r = l
and mention how the r > l case has already been handled
in part a
maybe note this
something like
if r >= l then r > l or r = l. if r > l then by part a it follows that there exists n_0 such that a_n < r whenever n >= n_0 so a_n <= r
and then can you try the r = l case
what am I supposed to say about d)?
can you show me your solution for the r = l case first
r != l
oops, ok
you good?
no
where are you stuck
r = l
oh, you mean I need a counterexample
that works
right
i was trying to show it was true
oh
an < r => an <= r?
because < is more strict
and because a <= b is by definition a < b or a = b
and in an OR we just need one of them
i already wrote out the r > l case
yes
i translated the symbols into something concrete which i knew wasnât true
for r = l is it possible to assume its true and get to an absurdity
sure
h0w
so the statement becomes if r = l then there exists n_0 such that a_n <= r = l for all n >= n_0
where a_n â> l
a subtlety here is that a_n is an arbitrary sequence with limit l
so itâs really saying for all sequences a_n that converge to l, this holds
to contradict it youd need to find a single sequence with limit l so that for every n_0 there is some n >= n_0 such that a_n > l
take a_n = l + 1/n
the contradiction is it doesnât hold for every sequence which is the first condition we had
this has limit l but a_n > l for every n in fact
do you follow?
i see
for every n there exists some n0
i negated the statement
weâre trying to show itâs false
so show its negation is true
p -> q = not(q) -> p
you mean not p
i didnât mention that but sure
you are not using it?
i mean contradiction and contrapositive are much of the same
just a difference in wording usually
why you negated
^
proving the negation is true is the same as proving the original is false
right thats contrapositive
goddamn
mrbeast
fuck that guy
đ¤
maybe itâs easier to see with symbols
forall n in N there exists some n0 in N negated is forall n0 in N there exists some n in N?
$\lnot(\forall (a_n){n \in \mathbb{N}} : a_n \to \ell, \exists n_0 \in \mathbb{N} ,, \forall n \geq n_0, ,, a_n \leq \ell) \iff \exists (a_n){n \in \mathbb{N}} : a_n \to \ell ,, \forall n_0 \in \mathbb{N} ,, \exists n \geq n_0 : a_n > \ell$
knief
i shee
seems hard
itâs not so bad, just find a sequence that converges to l from above
this converges to l but is not upperly bounded by l
wdym from above
like the sequence members are coming from above l
they shrink towards l but they are on the right
so not bounded above by l
so l is a lower bound
what
but
yes i did make it a lower bound
iâm just saying this wasnât our only option
it could shrink around l from both sides
something like
a_n = l + (-1)^n/n
l still isnât an upper bound here
but itâs also not a lower bound
now the sequence is approaching l from both sides
but whatâs important is that it isnât only coming from below
this is what the question was asking
the original question was asking if a_n â> l, does l have to be an eventual upper bound for a_n
we found a sequence that showed this wasnât the case
namely a sequence that converges to l without l being an eventual upper bound
you can do this either by having l be a lower bound or by l being neither a lower bound or an upper bound
I see
if l is a lower bound then all of my sequence members are coming from the right of l
if l is neither a lower bound or an upper bound then i have sequence members getting closer to l on both sides
i gave examples of both
in both cases l isnât an upper bound
but l = r
yea
you can use l and r interchangeably since we are working under the assumption that l = r
that was the case we have been focusing on
an <= l for all n in N means l is upper bound of an
we know an -> l we just need to find a sequence that has limit l but that l is not an upper bound
is there some n_0 so that for all n afterwards, l is an upper bound
yes
but take for example an = l + 1/n
take n0 = 2
then forall n in N such that n >= 2
it is true that l is an upper bound and that an converges to l
@severe pond sussy
how is it an upper bound
my bad
yep
can you help with d
alright
we spent like 30 min or more talking about it, until I finally understood c)
thatâs ok
ok so if r is an eventual strict upper bound for a_n what can we say about l
in relation to r
there are three possibilities to consider
can you guess what those might be
why?
I am trying to verbalize it
suppose r = l, then an -> r and we know that an -> r and r being a upper bound is not valid for all sequences
this is true by part a
yes
hmm
it doesnât have to necessarily be true
is it ever true
we donât care if itâs always true, just if itâs ever true
is it possible that l = r
ok the better way to say it is, is it possible that l is an eventual upper bound for a_n
iâm using l and r interchangeably here assuming l = r
my intuition from the drawing tells me no, because we can have sequences that get outside of the radius
but, idk
i think youâre misunderstanding what the question is asking
itâs asking for the set of possible l such that that condition holds for some sequence converging to l
we know that every l < r works because of part a
regardless of the sequence a_n
if l = r there are in fact sequences a_n for which this holds
yes for example the constant sequence an = c, c is an upper bound aswell as a lower bound of an
well we want a strict upper bound
we want a_n < l
so like
take a_n = l - 1/n
l is a strict upper bound
and the limit of a_n
so this works
right
so itâs possible that r is the limit of a_n here
all we know about r is that itâs an eventual strict upper bound
we are asking if l can equal r
i.e. can r be the limit of a_n
we just answered it
it can
itâs possible
so can l = r
and l can definitely be less than r
so l <= r
what is the difference between c) and d)
can l be greater than r?
c is asking if l <= r implies r is an eventual upper bound for a_n. d is asking for a condition on l if r is an eventual strict upper bound for a_n
dude
if l > r and r is an strict upper bound
then l is another strict upper bound
you are asking if its possible that an -> l aswell as l is an strict upper bound
yes it is possible
take an = l - 1/n
this converges to l and l is an strict upper bound
hmm
but youâre missing something
r is a strict upper bound that is less than l
if r is below l while also being a strict upper bound, how can a_n get close enough to l to converge to l
don't get it
imagine a number line
r is some point to the left of l
and every sequence member is below r
can a_n converge to l?
or is the sequence too far away
no, it converges to r
not necessarily
letâs say a_n = 1/n, r = 2, l = 3
clearly this wonât work
my points need to cluster around l
since r < l there is some fixed distance from my sequence to l
i have a_n < r < l
i can never get in between r and l
so i can never get close enough to l to converge to l
if l > r and an -> l then r is never an upper bound
well it might be an upper bound for finitely many of the terms in the beginning
itâs just not an eventual strict upper bound
it can never be an upper bound for the tail
how to formalize
so we already established that l <= r
now we are showing itâs impossible for l > r
ok
so assume l > r
we are trying to contradict l being the limit of the sequence while r being an eventual strict upper bound
since r is an eventual strict upper bound there is some n_0 such that a_n < r for all n >= n_0. but a_n â> l so there exists N >= n_0 such that
|a_n - l| < l - r whenever n >= N. hence
r - l < a_n - l < l - r <â> r < a_n < 2l - r
contradiction
since a_n < r
btw i said there exists N >= n_0 here because we know first that there is some N so that for all n >= N, |a_n - l| < l - r from the limit definition but i also want N >= n_0 to contradict the eventual upper bound so i can just take the max of N and n_0
if N was less than or equal n_0 then choose n_0 as N
nice!!
else, keep N
is it?
what?
n0
yes but we already used n_0 for a label
i specified that i wanted N to be >= n_0 where n_0 comes from r being an eventual upper bound
itâs subtle but important
ok i see
i did that because i wanted both conditions to hold simultaneously
i wanted both r to be an upper bound for those n and for a_n to be within that distance from l
but clearly they couldnât both be true
since i got contradicting inequalities
so our assumption that l > r was wrong
it can never be true
hence
l <= r
why do you need an n0 for the strict upper bound, there exists some n0? like if r is an strict upper bound its that an < r forall n in N
i shee
so i wanted to choose my N >= n_0 as well
so that i get this
so that a_n is both bounded above by r and within a distance l - r from l
would it be easier to add an extra step where you make K = max(n_0, N)
then take n >= K
itâs the same thing
and have no condition on N other than N in N
i shee
jajaj
math really is nasty
im sorry
what for
is great when you understand it but painful when you not, ykwim
I appreciate the help, sorry for making u waste so much time
yes
hopefully you got something out of it
i did learnt a lot today, though this shit is tough
I was having an easy time with the exercises 1 to 10
exercises 10 and 11 stuff started to get complicated real fast
i see
this exercises made it look like my linear algebra questions were easy
shit definitely started to get real very fast
not surprising
there are a lot of moving parts you need to keep track of in analysis
cya, tyvm for your help sir
very precise
youâre welcome
.solved
Closed by @spiral saddle
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
⢠Ask your math question in a clear, concise manner.
⢠Show your work, and if possible, explain where you are stuck.
⢠Do not immediately ping people or roles. After 15 minutes, feel free to ping <@&286206848099549185> once.
⢠Type the command .close to free the channel when you're done.
⢠Be polite and have a nice day!
Read #âhow-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
Channel closed due to the original message being deleted.
If you did not intend to do this, please open a new help channel,
as this action is irreversible, and this channel may abruptly lock.

Send your question here to claim the channel.
Remember:
⢠Ask your math question in a clear, concise manner.
⢠Show your work, and if possible, explain where you are stuck.
⢠Do not immediately ping people or roles. After 15 minutes, feel free to ping <@&286206848099549185> once.
⢠Type the command .close to free the channel when you're done.
⢠Be polite and have a nice day!
Read #âhow-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
Please don't occupy multiple help channels.
Channel closed due to the original message being deleted.
If you did not intend to do this, please open a new help channel,
as this action is irreversible, and this channel may abruptly lock.
Send your question here to claim the channel.
Remember:
⢠Ask your math question in a clear, concise manner.
⢠Show your work, and if possible, explain where you are stuck.
⢠Do not immediately ping people or roles. After 15 minutes, feel free to ping <@&286206848099549185> once.
⢠Type the command .close to free the channel when you're done.
⢠Be polite and have a nice day!
Read #âhow-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
Please don't occupy multiple help channels.
Channel closed due to the original message being deleted.
If you did not intend to do this, please open a new help channel,
as this action is irreversible, and this channel may abruptly lock.
Send your question here to claim the channel.
Remember:
⢠Ask your math question in a clear, concise manner.
⢠Show your work, and if possible, explain where you are stuck.
⢠Do not immediately ping people or roles. After 15 minutes, feel free to ping <@&286206848099549185> once.
⢠Type the command .close to free the channel when you're done.
⢠Be polite and have a nice day!
Read #âhow-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
hi, want to double check my working with someone, all the notations and stuff too please
thanks!
saying that a sum is equal to the limit of its terms is wrong
you should say you're using the test for divergence and then show you computing the limit
and as always, writing out arithmetic with infinity is abusive as well
the rest is fine
i see, so absuive as in not recommended or illegal in maths?
good now?
Not good practice. Some profs may flare up at the use of it.
i see, so should i just skip that step and just right 3 + 0 + 0 ?
like in the next one?
gotcha
^
you may want to include $\lim_{x\to \inf} 1/x$ instead of $1/\inf$
Annie Maqionde
yup all good 
Closed by @twilit cargo
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
⢠Ask your math question in a clear, concise manner.
⢠Show your work, and if possible, explain where you are stuck.
⢠Do not immediately ping people or roles. After 15 minutes, feel free to ping <@&286206848099549185> once.
⢠Type the command .close to free the channel when you're done.
⢠Be polite and have a nice day!
Read #âhow-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
How is it that $x^n = \overbar{a(x)b(x)}$
Wai
Compile Error! Click the
reaction for more information.
(You may edit your message to recompile.)
isn''t 1 in P too
here ?
yes
no 1 isn't in P else it would be the whole ring by multiplicativity
one more thing
Ok by hypothesis all the coef of f are in P except the first one
let a_0 and b_0 be constant terms of a and b
oops, right
and as 1 not in P that is possible
makes sense
tysm again
Yeah 1 is not in P because a prime ideal is by definition an ideal â (1) such that ...
.close
Closed by @desert pasture
Use .reopen if this was a mistake.
Send your question here to claim the channel.
Remember:
⢠Ask your math question in a clear, concise manner.
⢠Show your work, and if possible, explain where you are stuck.
⢠Do not immediately ping people or roles. After 15 minutes, feel free to ping <@&286206848099549185> once.
⢠Type the command .close to free the channel when you're done.
⢠Be polite and have a nice day!
Read #âhow-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
I heard in this YouTube video that all parabolas are similar but I didn't really understand the proof.
I want to start by creating a definition of similarity that uses the coordinate plane.
,tex the two functions $f$ and $g$ are similar iff $\\forall x,\frac{f(x)}{x}=\frac{g(x)}{x}$
bored amogi
Does this work? Why or why not?
can you not express f and g in terms of ax2+bx+c and dx2+ex+f?
Well
yea
and then divide by x and show that they're equals up to a constnat
First of allyour definition doesnt work for x=0
oh thats true
i got no clue
it was up to a constant now it's equals
The definition should just be
and then even if its up to a constant not all parabolas are 'similar'
isnt this supposed to be fx=kgx
f(x) = k g(x), for all x!=0
because you can ensure a=kd but then what about b=ke?
idk tbh
Dividing by x makes no sense
he was saying something about the 'aspect ratio'
which i imagine would be y/x
I think what you're looking for is:
suppose f and g are 2 parabolas
Tf=g for all f,g belonging to the set P2(F)
(set of degree 2 polynomials over a field)
and T is a linear transformation
?
whats a 'linear transformation'?
oh a transformation that preserves linearity right
uh maybe? im not sure
lmao idek what your question is
he did say that all circles are similar, maybe we can start with that instead?
the question looks simple enough what is this linearity thing lmao
if f(x)=4x^2+6x equals to g(x) if and only if f(x)= Kg(x) then k is a whole number and it should look something like 2g(x)=2(2x^2+3x) is it not?
why not x=0?
um, f(x)=4x^2+6x=g(x) iff f(x)=kg(x) is false?
unless k=1
"There exists a real number k such that f(x) = k g(x) for all x in the domain"
that wouldnt work--translation is an isometry and translation isnt covered under that definition
if k is one it is false and shouldn't it be different and special for every function?
could you prove how thats false?
oh sorry i forgot its f(x)=4x^2+6x=g(x)
or am i misunderstanding you?
Yeah i think you re right
Let me see
because f(x) is not equal to g(x) for the specific f(x) we are dealing with for k=1, k can have values specific for g(x) u are dealing with when multiplied altogether u should be getting fx . hey i think what u are looking for is something more rigorous and maybe i would not know that, what are u working on ?