#help-49
1 messages · Page 67 of 1
What was this question for?
before beginning though we need to take care of the worst case scenario for $1/|b_n| \leq M \implies |b_n| \geq M' > 0$
[ \lim \f{a_n}{b_n} = \f ab \quad b \neq 0 ]
neon
if both a_n and b_n are convergent sequences
hope this cleared that up
Bro's cooking 🗣️🗣️
neon
Oh. So you want to prove that, if $\lim a_n=a$ and $\lim b_n=b$, then $\lim \frac{a_n}{b_n} = \frac ab$?
SWR
yes
Okay. Let me read what you've done so far. You going hella fast rn'
yeah sorry about that
im spewing my thoughts
before I write it down
Yeah. This is what you want to do.
Then you can use product of limits, which is (slightly) easier than quotient
Yeah. It's annoying
I don't remember the exact strat, but I have it saved somewhere. Let me go look and I can at least make sure you end up going in the right direction.
Okay, thanks for that
What's the question 😭😭?? Congrats on solving lol
This is the question
SWR wrote it up there properly
Okay yeah. It's a bit subtle how to go forward. Just go as far as you can and ping me if you need advice.
Ohhhhh thanks
You have it good here, yeah
there's an extra step you need here
Oh?
limit composition
Will I figure I need it if I keep going
Can you use continuous functions at all?
nope
hmm
okay I need to cook too then.
I did this with the continuous function 1/x, and then limit composition. I thought it would translate nicely to sequence limits.
But yeah I need to think more a bit too
I mean end of the day I need to do something like:
[ |b_n - b| \f{1}{M'|b|} = \epsilon ]
neon
so $|b_n - b| = M'|b|\epsilon$
neon
But M' depends on n, right?
Can't I just let that be my epsilon
Yeah well it does
I need a sufficiently large M'
then it can't be your epsilon yet. You need a bound for M' independent of n
yeah
I did it with 1/x by using a reverse triangle inequality, but I don't know how to translate that here
I see
😭
good luck
@surreal moon couldn't we consider M' = |b|/k where k is an arbitrarily large positive number
yes, but you need to find k now
hmm
is "arbitrarily large" not good enough a claim to fight in court
It should be fine except for if there's a random 0 in the line
no it is not
hmm alright
@surreal moon what if we looked for a sufficiently large n
would that be good
that's what we are trying to do
What do you mean?
okay lets just say k = 3 for now
for some value of n say N_1
|b - b_n| < |b|/3
is that alright to say
should be actually yeah
there isn't a problem
this is regular old epsilon-N
our epsilon is just |b|/3
hold on I'm having trouble understanding what you are trying to say. Lemme take a deeper look
What is k now?
no but like, what does k represent?
uhh
just a positive real
Am I struggling and dragging this out more than usual for this proof 
hard to say. Your proof is kind of fragmented so far, so it's hard for me to follow along. Would it be alright if you wrote out everything you had so far?
i'm struggling to understand why you can't just use multiply or divide the limits since they are both convergent
(perhaps just a SS of where you are actually working on this)
that's what we're proving
You need to prove that multiplying works through limits. That's the task
Okay
oh okay
wait I think I got it
articulation in process
We aim to prove that given $\lim a_n = a$ and $\lim b_n = b \neq 0$, $\lim a_n/b_n = a/b$. If it is proven that $(b_n) \to b \implies (b_n^{-1}) \to b^{-1}$, then the rest of the proof follows from an established statement. Consider:
[ \left | \f{1}{b_n} - \f 1n \right | = \f{|b_n - b|}{|b||b_n|} ]
We wish to accomodate for the boundedness of $1/|b|_n$ requiring an $M'$ such that $|b_n| \geq M' > 0$. Consider for some $n \geq N_1 \in \set N$, $|b_n - b| < |b|/2$. With the triangle inequality, we can conclude that $|b_n| > |b|/2$. Next, for suitable $n \geq N_2 \in \set N$, we have:
[ |b_n - b| = \epsilon \cdot \f{|b|}{2} \cdot |b| ]
Piecing everything together, we have:
[ \left | \f{1}{b_n} - \f 1b \right | = \f{|b - b_n|}{|b|(|b|/2)} < \f{\epsilon |b|^2}{2} \cdot \f{1}{|b|(|b|/2)} = \epsilon ]
neon
I assume you mean -1/b here?
oh yeah
So you want to find an $M'$ (which depends only on $\epsilon$ I assume?) that serves as an upper bound for $|b_n|$ for sufficiently large $n$?
SWR
I mean my M' here is |b|/2 it doesn't depend on epsilon
oh ok
the real jump was going from |b_n - b| to |b_n|
because that satisfies the condition on M'
So wait, you're saying that $M'=\frac{|b|}{2}$?
SWR
And you're saying that $|b_n|\ge\frac{|b|}{2}$ for sufficiently large $n$?
also you can modify this bit to whatever we want to accomodate the M'
SWR
yeah
if we did |b|/3 then this bit would become 2|b|/3 I guess
and here we have 2|b|/3 instead of |b|/2
so it still all cancels back to epsilon
is it satisfactory?
phew
and it was cleaner than mine
¯_(ツ)_/¯

I'll get on your level soon

@twilit field in case you were wondering you're not alone in the whole proof taking a lot of time department 
Alright then, thanks again SWR, I'll head to bed now
.close
Closed by @sage helm
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
!show
Show your work, and if possible, explain where you are stuck.
.close
Closed by @grave locust
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
$\phi$ is the Number-Theoretic Euler's function
Halex
@meager ore Has your question been resolved?
@meager ore Has your question been resolved?
is this question is poorly written. i believe the a_i's are supposed to be relatively prime to n
not just any phi(n) residues
what do you mean with "not just any phi(n) residues"?
like... take n = 4. then phi(n) = 2 but letting a_1 = 0 and a_2 = 1 makes a_1 + a_2 not 0 mod n
you cannot just take any 2 residues
they should be distinct, and the ones relatively prime to 4 (so, 1 and 3)
so i'm just pointing out the problem is unclear with that but that's probably what was intended
but anyway for the problem, maybe you can think about pairing up the a_i's so that they add up to 0 (mod n)
you may want to look at example (say, n = 20) to see what's going on
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
So I'm trying to prove gcd(a,b) can be expressed as ax+by
where x an y are integers
so my book started off with the claim the $S= {au+bv|au+bv>0; u,v \text { are integers }}$
ƒ(Why am. I here)=I don't Know
and from this it claims d=ax+by
Ah good ol’ bezout’s identity
well you skipped over so much
I’m fairly certain this comes directly from the division algorithm
we dont know what your book covered until then
ok, sorry
got it
thanks
Just write the division algorithm backwards
so $x=aq+d$
ƒ(Why am. I here)=I don't Know
dont mix up division algorithm and euclidean algorithm
(When you say "division algorithm", that's the same thing as the Euclidean algorithm, right?)
the book here uses wop and not euclidean algorithm
ah, my bad
the book calls the result that division with remainder exists the division algorithm
oh okay
sorry, I meant the Euclidian algorithm
you dont need the euclidean algorithm for the proof
hmm? there's another proof?
yeah consider the set from above and use wop. or wait a min until they send the proof
wop?
well ordering principle?
yes
oh I think I understand how
||considering the minimum element and proving it's the GCD?||
yes
nice
ok, so now what exactly is your question
d is in S. by definition of S that means d=ax+by for some integers x,y
that was my question, d is the divisor, right, how can we claim it will be of the form ax+by
at that point we havent said anything about d being a divisor
Closed by @twilit field
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
BGD=BCD-BGC-CGD
no?
i mean they are both the same ig
alr do you know what the length of the side of the square is
can you simplify it
uh
i meant like 32=16×2 and take the 16 out
ye
alr, do you know how to find area BCD?
yes
the next easiest is finding CGD
if we let CD be the base, do you know the height of the triangle?
which is?
would you say the distance from the base to G to be half of BC?
since G is the midpoint of F to C and the height of F to the base is the same as BC, so G would be half of BC
if you manage to get the area of CGD, you can do the same process for BGC
what
do you understand that if GC is half of FC, basically the height from F to CD is half of the G to CD
i mean you can prove this with simmilar triangles ig
sait your asking q2 right
blue and yellow are simmilar
so like, the height of blue/hyp of blue=height of yellow/hyp of yellow
hyp of yellow is 2 hyp of blue
so height of yellow would mean 2 times height of blue
huh
uh
try bc and imagine the midpoint, now try moving point b horizontally, if you think about it, the height of the midpoint woulf still be the same just that the horizontal distance changes
why would they match up? BD is the hypotenuse
heres me graphing it
the one on tbe black line the midpoint and the one on the red line is the B when you move it, and 0,0 is C
you can see that the midpoint still sits at the black line
hopefully you can unferstand my bad teaching 🙏
do you get it? like hoe the height is half of BC
i dont understand what your asking
yea i get that but what are you asking
are you askibg why they are diffrent?
wdym meet up??
btw what grade of math is this?
what
what grade are you
oh oj
have you been g9 for a while or just got to g9?
ooo same
can i ask what you mean with "meet up"?
huh
like why if its half of the hypotenuse it would also be half of the veritcal thing
.
i think this might help
ignore the left side of the figure
wait let me draw it for you
half of the hypotenuse means if one of them is at (0,0), the midpoint of the hypotenuse would be (x/2,y/2) if the hypotenuse is (x,y)
im not really sure how to explain it
what shape is only this part
yes good
now what is line FC
1 sec lemme show u more clearly
what is FC of the rectangle
exactly
now if you get the midpoint of the diagonal of the rectangle....
it's the midpoint of the rectangle!
and that is point G
therefore the verticles are equal
oh no it's completely fine
umm do you know properties of a rectangle
"diagonals bisect eachother"
the point where both diagonals meet
what bisection means is that it divides the line into 2 equal parts
think of the centre of a plus sign
those 2 lines bisect eachother
meaning the centre of the line
yeppers
that meeting point is not only the centre of the rectangle
but also the centre of each diagonal
lemme show u what i mean
AX = XC
BX = XD
i don't understand what you mean by proof
give me a second here i'll give you something a little more concrete
the strongest proof is just construction
if you go and plot a rectangle on a graph and draw its diagonals, you'll see that the intersection point divides the line into 2 equal parts
what do you want
an algebraic proof?
what proof do you want
with similarity or something?
i mean
keep this figure in mind
we know it's a rectangle right
AB = CD, AD = BC
so we can prove congruency
use side angle angle
AXB is congruent to CXD
AB = CD opposite sides of a rectangle
and ABX = CDX by alternate angles, AB || CD
BAX = DCX
after proving congruency you can assume AX = CX
BX = DX
do the same for triangles AXD and CXB
and voila
ye
huh
uhh did you understand how the height of CGD is half of BC now?
alright, so whats the height of the triangle?
then whats the area
well BC to G is half of BF, and BF is half of AB
and you know AB
use the same logic as how you get height of CDG as half of BC, instead of half of BC, its half of BF
alr so whats BF?
so whats the height of the yriangle then
@last slate Has your question been resolved?
BCG
now use this
@last slate Has your question been resolved?
Closed by @fresh blade
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
if $a|b$ and $c|d$ then show $ac|bd$
ƒ(Why am. I here)=I don't Know
so $b=ka$
ƒ(Why am. I here)=I don't Know
$d=k'c$
ƒ(Why am. I here)=I don't Know
so so $db =(k k ' ac$
ƒ(Why am. I here)=I don't Know
so $db = Kac$
ƒ(Why am. I here)=I don't Know
thus $ac|db$
ƒ(Why am. I here)=I don't Know
is that right
yes
Closed by @twilit field
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
Prove $8|5^{2n}+7$
ƒ(Why am. I here)=I don't Know
using induction
for $ n \geq 1$
Now I can solve this using the binomial theorm
For induction, these are my thoughts
yeah, not too sure of how I'd use induction here
becomes rather messy
no?
wait
I think I know a theorm that may help
you dont need any theorems here
just set up your induction as usual and use your inductive step to prove the n+1 case
so $8|25 \cdot 5^{2n}+7$
ƒ(Why am. I here)=I don't Know
using the fact that 8 | 5^2n + 7
its easier imo if you unpack what "divides" means
and use equations
am I allowed to do that in NT
use the definition of divides? yes
ok so $\frac{5^{2n}+7}{8} =k$
ƒ(Why am. I here)=I don't Know
okay but you should steer clear of division whenever possible
ƒ(Why am. I here)=I don't Know
ok, that doesn't seem to work
you never need to divide anything to do this
ƒ(Why am. I here)=I don't Know
I use this?
yep, and now youre trying to show that 5^2(n+1) + 7 is a multiple of 8 using that
so $7=8k-5^{2n}$
ƒ(Why am. I here)=I don't Know
ƒ(Why am. I here)=I don't Know
nice
.close
Closed by @twilit field
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
is this fine?
right idea, although you've not said what N is in your proof
yeah
go ahead
Abbott
Kinda hit a roadblock
The problem is that the quantity on the right is greater than epsilon
Full title? Understanding Analysis?
That's the one
So you've noticed essentially that [|\sqrt{x_n}-\sqrt x|=\frac{|x_n-x|}{|\sqrt{x_n}+\sqrt{x}|}]
might need to split cases
Edward II
Oh and that is lesser than replacing x_n with M
split cases how?
to ensure the denominator doesn't die
The denominator can't die
why would it die
We've assumed everything is >0

oh that's a good point
you need to split cases based on whether x = 0
Hmm
if it is, you've already done the proof
if it isn't, you can ensure the denom doesn't die
Right I can declare x isn't 0 here
yes
I mean alright but that doesn't adress my point of the inequality
well you see
Do I restructure around this
Hmm true
that's enough to save you
Okay let me write it in mathematician
oh
bruh
wait how is this supposed to help me then
It just kicks the can down the road
Okay that makes sense
@hard umbra sorry it took a while
The only thing I'm confused is if I need to do the N1 N2 thing
for epsilon' and the main proof statement
technically yes
And then choose max{N1, N2}
you'd need to justify why epsilon' exists
Alright, beyond that it's all good?
yes
was the justification in the () not enough
well i mean thats leading up to why
but to get into the weeds, you'd have to unpack the definition of convergence
👍
Closed by @sage helm
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
Prove that for any integer a, one of the integers, a,a+2,a+4 is divisible by 3
so I first assumed a is of the form 3n
in that case it's trivially divisible by 3
then in the from 3n+1
in which case 3n+3 will be divisible by 3
and lastly 3n+2
in which case 3n+6 is divisible by3
would this be enough
or is it lacking rigour
Just stipulate that every integer is of one of the forms 3n, 3n+1 or 3n+2
Rest is good
don't I have to prove that in. a NT class? 
I mean it makes sense
but do I have to prove that first?
There’s a theorem that says that
I mean, you can
I think*
hmm
Let me think of how I'd prove it
I guess I could prove that a number n, and it's successor are always co prime
But that doesn't help much

what happens if you add 1 to 3n+2
Yeah a proof by induction is what I was thinking
i like how the sort of thinking youd use when using mod 3 is sort of already intuitive to @twilit field now, without learning it yet
Thanks!
Yeah, modular arithmetic is gonna come really naturally to you
I hope so, I need a perfect score in my UG
You’re forming the base ideas of it naturally
thanks
oh and youre doing all this before even starting, yea it will go smoothly for you!
oh you need a perfect score? I honestly never think its that necessary unless you maybe plan on doing a PhD or something
you have the division algorithm. use that to prove this claim. no need to use induction explicitly. (tho it wouldnt be bad to practice it of course)
yeah division algorithm also works
i'm not studying at a great college,it's good, but that; about it, so i need that for my msc
hmm, thanks
.close
Closed by @twilit field
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
How to factorise x³-6x²+11x-6
!status
What step are you on?
1. I don't know where to begin.
2. I have begun but got stuck midway.
3. I got an answer but I was told that it's wrong.
4. I got an answer and would like my work checked.
5. I have a question about someone else's work/solution.
6. I have completed the problem and don't need help anymore. Thank you.
7. None of the above
try finding one root first
1
I want to learn how to factorise these type of question
$\frac{x^3-6x^2+11x-6}{x-1}$
Flappie
Ok
not necessary to do long division here, you can do it by using sum and product of roots
How sir pls teach me
the original equation has roots whose sum equals 6
How do we know that can you pls tell from basic
I mean about that roots
What is that
you try lil numbers like 1, 2, 3 or -1, -2, -3 and check if it's equal to zero
if you have a polynomial that goes like x^n.... + k
you're not forced to check through all "small" numbers to find the root
the rational root theorem tells you that in those cases, only divisors of k can be integer roots
so +/- 1, +/- k, etc...
Ok
5
in our case
Yes
we have all divisors of 6, plus or minus
so 1,2,3,6 or -1,...
we found 1 so we're good to go
then we can factor
well
factor this by (x-1)
and then you're left with a quadratic
In this case it should be 6??
-> quadratic formula or whatever you prefer
?
That root thing??
I said the integer roots can only be 1,2,3,6 or -1,-2,-3,-6
we never said
all roots are in here
x^3 -6x^2+11x-6 = (x-1)(ax^2 +bx + c) you can do like this too
and we found that "1" is a root
so we can factor exactly like this
and then you're left to factor ax^2+bx+c
Like x is a polynomial and the value of x is 1
again, it's not necessary to factor like this
factoring using Viete's relations is quite faster
What I am reading things like viete and quadrictic equation is going over my head
Pls tell for a class 8 boy
you probably do know what viete's relations are, just not by the name
if you have a quadratic $ax^2+bx+c=0$ can you tell me the sum of its roots?
kheerii
1???
I didn't really understand what is (sum of roots)
Pls explain one more time
??
Sir??
. Close
.close
Closed by @shadow sparrow
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
When doing integration by parts how do we decide which term will be “u” and “dv”
Do we just pick at random
just think about which one is easier to integrate and which one is easier to differentiate
there's also an acronym you can use to help you but it's not always correct
L: logarithmic
I: inverse trigonometric functions
A: algebraic (polynomials)
T: trigonometric
E: exponential
These tell you in which order you should choose your u
L and I are interchangeable
Note that this isn’t a hard and fast rule, it’s just a guide
Ok thx
You will find instances where this isn’t the best approach
This is a general rule
.close
Closed by @drowsy thunder
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
prove that the difference if two consecutive cubes is never divisible by 2
what have you tried so far
ƒ(Why am. I here)=I don't Know
ƒ(Why am. I here)=I don't Know
Closed by @twilit field
Use .reopen if this was a mistake.
If you want a simpler argument: n^3 is always the same parity (odd or even) as n. So since n+1 and n are opposite parities, (n+1)^3 and n^3 will also be opposite parities. Therefore, their difference is odd because even - odd or odd - even are always odd.
(I left some details out about how you prove the claims I made, but that's the gist)
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
Prove the Lax-Milgram lemma:
Let \( a : X \times X \to \mathbb{C} \) be a bounded coercive sesquilinear form on the Hilbert space \( X \). Then there exists a bijective linear bounded operator \( A : X \to X \) such that for all \( x, y \in X \),
\[ a(y, Ax) = (y, x)_X. \]
Hint: According to the Riesz representation theorem, there exists a unique element \( Ax \) for each \( x \) satisfying \( a(y, x) = (y, Ax)_X \) for all \( y \in X \). Show sequentially: (i) \( N(A) = \{0\} \), (ii) the image space \( R(A) \) is closed, and (iii) \( R(A) = X \).
tobi
I already managed to proof that the kernel of A is trivial.
N denoted the kernel and R denotes the Imagine of A (I dont know why they use this notation in functional analysis its confusing to me).
I Would need help with the 2nd step
You probably want to post this in #advanced-analysis instead of in this help channel
(N = null space and R = range, probably)
okay thanks
.close
Closed by @uneven sandal
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
Show that: (α,β,γ are random vectors) I am not sure what the first step is or how should I approach similar questions. you can use a,b,c
do you already know this inequality if there were only 2 vectors instead of 3?
yes
what do you do after wont you just get |a+b| + |c| <= |a| + |b| + |c|?
i still dont understand
|a+b| <= |a| + |b|
|a+b| + |c| <= |a| + |b| + |c|
i can only get to here what do i need to do after
isnt |a+b+c| different from |a+b| + |c|? i mean they equal different things right?
Yes, they are different
but how do they relate?
Hint: a+b+c = (a+b)+c
|a+b+c| <= |a+b|+|c|?
oh i got it now
|a+b+c| = |(a+b)+c| <= |a+b| + |c| right?
.close
Closed by @wanton trail
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
I'm trying to show that the definition of the covariant derivative of a tensor density implies that the covariant derivative of the Levi-Civita tensor density vanishes.
This definition is given:
$\nabla_c T^{a...}{b...} = \text{usual terms if $T^{a...}{b...}$ were a tensor} - W\Gamma^{d}{dc}T^{a...}{b...}$
For the Levi-Civita tensor $\epsilon^{abcd}$, I have:
$\nabla_e \epsilon^{abcd} = \partial_e \epsilon^{abcd} + \Gamma^{a}{fe}\epsilon^{fbcd} + \Gamma^{b}{fe}\epsilon^{afcd} + \Gamma^{c}{fe}\epsilon^{abfd} + \Gamma^{d}{fe}\epsilon^{abcf} - \Gamma^{f}{fe}\epsilon^{abcd} $
The middle terms cancel out due to the antisymmetric properties of the LC tensor density, so that I have:
$\nabla_e \epsilon^{abcd} = \partial_e \epsilon^{abcd} - \Gamma^{f}{fe}\epsilon^{abcd} $
The LC tensor density is just a generalisation of the Dirac delta function which vanishes under partial derivatives, leaving:
$\nabla_e \epsilon^{abcd} = - \Gamma^{f}_{fe}\epsilon^{abcd} $
But I'm not sure how to argue that the last term must be zero. I think one could argue that it's to do with the connection being symmetric and the LC tensor density being antisymmetric, but they don't have the same rank, so I'm not sure if that follows, exactly.
momoji
Oh, sorry about the poor formatting. I can try to make it a little clearer.
Objective: I'm trying to show that the definition of the covariant derivative of a tensor density implies that the covariant derivative of the Levi-Civita tensor density vanishes.\newline
The definition given:\newline\newline
$\nablac T^{a...}{b...} = \text{usual terms if $T^{a...}{b...}$ were a tensor} - W\Gamma^{d}{dc}T^{a...}{b...}$
\newline\newline
For the Levi-Civita tensor $\epsilon^{abcd}$, I have: \newline
$\nabla_e \epsilon^{abcd} = \partial_e \epsilon^{abcd} + \Gamma^{a}{fe}\epsilon^{fbcd} + \Gamma^{b}{fe}\epsilon^{afcd} + \Gamma^{c}{fe}\epsilon^{abfd} + \Gamma^{d}{fe}\epsilon^{abcf} - \Gamma^{f}{fe}\epsilon^{abcd} $
\newline\newline
The middle terms cancel out due to the antisymmetric properties of the LC tensor density, so that I have:
$\nabla_e \epsilon^{abcd} = \partial_e \epsilon^{abcd} - \Gamma^{f}{fe}\epsilon^{abcd} $
\newline\newline
The LC tensor density is just a generalisation of the Dirac delta function which vanishes under partial derivatives, leaving:\newline
$\nabla_e \epsilon^{abcd} = - \Gamma^{f}_{fe}\epsilon^{abcd} $
\newline\newline
But I'm not sure how to argue that the last term must be zero. I think one could argue that it's to do with the connection being symmetric and the LC tensor density being antisymmetric, but they don't have the same rank, so I'm not sure if that follows, exactly.
momoji
Compile Error! Click the
reaction for more information.
(You may edit your message to recompile.)
(Hopefully that's readable aha)
Oh, some additional notes, I suppose. There's a derivation of this online that uses the metric compatibility of the LC tensor to get to the end result. But my textbook (D'Inverno) seems pretty explicit about using the definition given.
@mystic elk Has your question been resolved?
@mystic elk Has your question been resolved?
@mystic elk Has your question been resolved?
@mystic elk Has your question been resolved?
(yup, that's right!)
wait what does the W mean in your definition of covariant derivative?
also you might get better help in #diff-geo-diff-top than in this help channel
It's the weight of the tensor density. In the case of the Levi-Civita tensor density with all contravariant indices, it's +1
Ahh okay, I thought maybe I should've posted it there haha.
@mystic elk Has your question been resolved?
.close
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
Prove that $1 \cdot x = x \forall x \in F^n$
ƒ(Why am. I here)=I don't Know
so I thought of starting my adding the additive -1(x) to both sides
$1(x)-1(x)=x-1(x)$
ƒ(Why am. I here)=I don't Know
so $1(x-x)=x-1(x)$
ƒ(Why am. I here)=I don't Know
ƒ(Why am. I here)=I don't Know
is this right?
how about actually just doing the multiplication?
I dont know if your axioms allow you to factor the 1 out like you did already
I have to prove that (1)x=x though
1(x) is just x
thats what you are supposed to show
use the definition of scalar multiplication in F^n to compute 1*x
like x in F^n means you have some sense of x consisting of n scalars
$1(x_1,x_2,x_3....x_n)$
ƒ(Why am. I here)=I don't Know
which is $(x_1,x_2....x_n)$
ƒ(Why am. I here)=I don't Know
right
dont skip steps
$(1x_1,1x_2....1x_n)$
ƒ(Why am. I here)=I don't Know
and now this is equal to (x1,...xn) why?
1 is the multiplicative identity
in?
ƒ(Why am. I here)=I don't Know
well thats what you are supposed to show
here 1 is the multiplicative identity in F, which is why you can safely say... ?
in that case I'll probably use the fact that $1(a)'s additive inverse is -1(a)
so 1(a)-1(a)=a-(1)a
so a=1(a)
x in F^n?
I have to go now( have a class), can I close this for now?
yes
but you already tried that at the beginning
you're making it more complicated than it should be
you just come back to this
i mean yeah you can always close it. but heres what i was getting at: since 1 is the multiplicative identity in F, and since $x_1,\dots,x_n$ are all in F, you can safely say $1\cdot x_i = x_i$
and knowing 1 is multiplicative identity in F
esca (@ with reply)
why sorry
oh no worries lol. good luck
Closed by @twilit field
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
write it as a piecewise function and find the conditions for differentiability at x=0
@bleak pier Has your question been resolved?
Piecewise how?
@bleak pier piecewise means write it using cases:
[
f(x) = \begin{cases}
a_0 + a_1 x + a_2 x^2 + a_3 x^3, & \text{if } x \ge 0 \
a_0 + a_1 (-x) + a_2 (-x)^2 + a_3 (-x)^3, & \text{otherwise}
\end{cases}
]
OmnipotentEntity
@bleak pier the idea is if f(x) is differentiable at x=0 then lim h->0 (f(h) - f(0))/h exists.
In order for this limit to exist the limit h->0+ and limit h->0- must both exist and be equal
An easier way to express this is if we say f+(x) = a0 + a1 x + a2 x^2 + a3 x^3, and f-(x) is the version of the function with x replaced by (-x), then we want f+'(0) to be equal to f-'(0)
How can we select our ai to make this true?
For x>=0 f'(x)=a1+2a2x+3a3x^2+....
f'(x-)=-a1+2a2x-3a3x^2
a1=0 and negative terms should be 0?
2(3a3x^2+5a5x^4+.....)
Well
We only care about the value of f+'(0) and f-'(0)
We only need that these are equal to each other at the one point, other points do not matter
@bleak pier
So
f+'(0) = a1 + 2a2 (0) + 3a3 (0)^2 = a1
f-'(0) = -a1 + 2a2 (0) - 3a3 (0)^2 = -a1
So all we need is a1 = -a1 or a1 = 0
Oh right!
You also need that f+(0) = f-(0)
But that is fairly trivial, because it reduces to a tautology. a0 = a0
But yes, the only point we need to be concerned about is the point x = 0. Because it's just a normal polynomial at every other point.
@bleak pier Has your question been resolved?
Closed by @bleak pier
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
if a and b are odd integers prove that $8|a^2-b^2$
ƒ(Why am. I here)=I don't Know
hey there
!status
What step are you on?
1. I don't know where to begin.
2. I have begun but got stuck midway.
3. I got an answer but I was told that it's wrong.
4. I got an answer and would like my work checked.
5. I have a question about someone else's work/solution.
6. I have completed the problem and don't need help anymore. Thank you.
7. None of the above
$\frac{(a-b)(a+b)}{8}$
ƒ(Why am. I here)=I don't Know
ƒ(Why am. I here)=I don't Know
kheerii
yeah
consider the parities of both the numbers on top
in particular consider their difference
is it odd or even
ah
odd-even
yes, since their difference is odd
Closed by @twilit field
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
I'm not sure I understand how $F^{\infty}$ is a vector space
ƒ(Why am. I here)=I don't Know
What are the vector space axioms that you feel would not apply?
I didn't undestand the part where they said it's the set of all sequences of elements of F
Every infinite ordered list whose elements are all elements of F
as in a power set ia. set of sets
I mean you believe without problem that {(x1,...,xn): x_i in F} is a set, no?
yes
its not a set of sets
well and now instead of the objects being (x1,...,xn), they are (x1,x2,....)
what
ƒ(Why am. I here)=I don't Know
yes
but these two arent sets
Closed by @twilit field
Use .reopen if this was a mistake.
.reopen
✅
yes?
yes what?
@dreamy lichen was typing
Nothing
That was like 3 mins ago i think
Closed by @twilit field
Use .reopen if this was a mistake.
Sorry
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
any hints?
trying to find a non brute force way
Informatic science or no ?
Im not good in computer science, but i can give you a website where they give some helps if you want
Closed by @livid bramble
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
what is this y for
between the pi and the brackets
also i think you forgot ||to square it||
good point
that is not supposed to be there
well for +-9 right
but do i make the bound x = -9, 9
or 0,9
yeaa
that dont help me
ahh ok yeah
quandrat 1
would 0 to 9
@fresh sparrow
<@&286206848099549185>
is this integral right?
its saying my answer isnt right tho
@white loom Has your question been resolved?
still doesnt
Closed due to timeout
Use .reopen if this was a mistake.
.reopen
✅
<@&286206848099549185>
@white loom Has your question been resolved?
@white loom 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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
help me pls
@vast remnant Has your question been resolved?
<@&286206848099549185>
@vast remnant Has your question been resolved?
better to go through the congruent cases again dude
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
can someone guide me on how to do A
my previous answer for A was 24310 but its wrong
do you know what the difference between permutation and combination is?
the difference between A and B is exactly that
@wooden cape Has your question been resolved?
I found the answer
but thank you
@wooden cape Has your question been resolved?
Closed by @wooden cape
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
My ans is c^11/2 d8
should be 2(c^6)*(d^7)
Oh sirry i add not minus
Why is it 2(c^6)
Shouldnt jt be c^6 / 2
how are you getting division by 2
yeh, so like i asked, how are you getting division by 2
why is 2 in the denominator
yes, that would be correct for that problem
based on the values initially present
$\frac{8}{32} = \frac14$
ℝαμΩℕωⅤ
$\frac{14}{7}$ is NOT $\frac12$
ℝαμΩℕωⅤ
@steel cairn Has your question been resolved?
Closed by @steel cairn
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.
• After 15 minutes, feel free to ping <@&286206848099549185>.
• Type the command .close to free the channel when you're done.
• Be polite and have a nice day!
Read #❓how-to-get-help for further information on how to ask a good question, and about conduct in the question channels.
if $d|n$, then prove $2^d-1|2^n-1$


