#help-38
1 messages · Page 226 of 1
Oh nice, I’m from Mumbai
U sub kar, u = cos(x)
Actually it’s an odd function
So can just use symmetry argument
How you check its odd?
f(-x) = -f(x)
Ohh yeah
cause cos is odd
Closed by @lean kraken
Use .reopen if this was a mistake.
Nws
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 want to model Y ~ X where Y is a proportion of groups (17 in total) that an observation belongs to. 2/17, 5/17 8/17, etc.
So it goes from 0-1 in steps of 0.06.
X is a normally distributed continuous variable.
Two questions
- Would it be incorrect to multiply the proportion by 17 to get the group count and do Poisson regression, given it it bounded on the right hand side of the distribution to 17?
- What would be the most appropriate way to model Y scaled as a proportion, given it isn't a continuous proportion?
@grizzled pendant Has your question been resolved?
<@&286206848099549185>
@grizzled pendant Has your question been resolved?
what do you mean by model? what parameters are you estimating? the mean/std of the normal random variable?
do you observe X or Y?
Regression model.
Y ~ X = Y independent variable, X dependent variable.
so you observe both?
Yes - I have Y and X values for each observation
The regression coefficient, and P-value, for the X variable
regression coefficient of what?
you're saying that Y is some discretized version of the quantiles of X
but there's nothing random about Y from X
I'm not sure you understand what I'm asking
I don't
Y ~ X doesn't mean Y is 'a discretized version of the quantiles of X'
Y ~ X means you are regressing Y against X. As in performing regression to see how X predicts Y (if at all)
I know that
but then what is all of this stuff about "proportion of groups" an observation belongs to? seems like you're saying that Y is some binned version of F(X), where F is the normal cdf
No?
My Y variable is the number of outcome groups for each observation, where the maximum number of outcomes is 17. Essentially it's a score out of 17 as the Y variable.
X is a different continuous normally distributed variable that goes from -1 to 1
you just said that X is normally distributed...
but okay
so you have 17 subsets of [-1, 1], and Y is the count of the number of subsets X falls into?
What???
Y is a score out of 17
X is a continuous variable that can be between -1 and 1
I want to know which is the best regression method for finding the association between Y and X (coefficients and P-values)
From what I understand Poisson regression would be incorrect because it assumes there is no upperband for the number of counts. In this case there is an upper band of 17/17
So I'm trying to figure out if using Y as a proportion (count/17) and using something like Beta regression would be more appropriate (with adjustment to deal with 0s and 1s)
It seems Binomial regression would be the most appropriate
@grizzled pendant Has your question been resolved?
@grizzled pendant Has your question been resolved?
<@&286206848099549185>
My current solution has been to use a Binomial model, where I give the Y variable as number of successes & number of failures. I'm still not certain if that is the most sound method though.
(interestingly, doing Poisson regression Y ~ X, and then Poisson regression of (Y*17) ~ X gives the exact same coefficients but completely different P-values. I didn't expect that)
@grizzled pendant Has your question been resolved?
@grizzled pendant Has your question been resolved?
Help please
!occupied
Someone else is already using this help channel. If you need help with a question, please open your own help channel/thread (see #❓how-to-get-help for instructions).
Hi! what exactly is it youre struggling with here?
Yes
And what is the radius in your picture?
10CM
5cm sorry
So 10 x 5 =50 and i must square it so it much be 250?
why 10x?
No 25 because 5 squared is 25
how many sides does your box have
"faces" *, not "sides"
(only because in this case "sides" is ambiguous)
My bad english is not my native language and this is my first time coming into contact with it in regards to math terminology
couldnt even remember what the shape is called
(I mean "cuboid" is in the question itself lol)
@grizzled pendant 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.
how do i prove: T:V->V linear trans.
0 is not an eigenvalue of T => T is invertible
finite dimensions?
do you see a relation to the kernel of T?
i want to prove without the limitation but if u can with then tell me how
i did that for the oppsite direction but i dont see how here
you need finite dim
what was the relation
then tell me how with it
what u see
huh
0 eigenspace = kernel
Tv = 0v = 0
so 0 not an eigenvalue means kernel = 0eigenspace = {0}
that only proves 1-to-1
in infinite dims there are injective linear maps which are not invertible
what i meant
in other words in infinite dims the claim is false
so you have if and only if in finite dimensions and in infinite dimensions you only have
L is invertible implies 0 is not an eigenvalue
it only works in finite dims
um yes this is true
but how do u prove it
example of a linear transformation that's injective and not invertible:
Consider the space of infinite sequences in R
(x1,x2,x3,...)
and the map
(x1,x2,x3,...) maps to (0,x1,x2,...)
try proving it using the rank-nullity theorem
yes
^
great, ty guys
<@&268886789983436800>
@fair forge 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.
I need help please someone (preferably French) but I really need someone to help me with a test that I have tomorrow
If you post your questions, someone will likely help (not me, but there are some French speaking people in the server)
Did you open a channel to ask about tomorrow?
Do you do some kind of pre-booking
I assume his test is tomorrow, which is why he's studying for it today.
@solemn pike 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.
I'm trynna show that this is not differentiable:
[
f(z) =
\begin{cases}
\frac{z^5}{|z|^4}, & \text{if } z \neq 0, \
0, & \text{if } z = 0.
\end{cases}
]
Can anyone think of a path I can take where this doesn't go to 0?
theaveragejoe6029
take the limits
@vapid pawn Has your question been resolved?
It does go to 0, but it doesn't do so in a smooth way
You'll want to apply the definition of the derivative at x = 0
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 im on part D right now
i have to make another vector that has the same relationship to t and u which was neither in part C
this is what i wrote so far
but i think i messed up
is this neither or parallel
i just doubled the values from t
@stoic garden are you still active
<@&286206848099549185>
If you do that
Why did you double it?
You just want a vector w such that w.t = u.t
No what???
He can
He just has to preserve the non-parallel / non-orthogonal relationship between the vectors
w cannot be a scalar multiple of t, that makes it parallel
Same relationship to t must surely mean that dot products must also be equal
Not necessarily according to the way the question is phrased in C
"Same relationship to t as it has with u"
Read C
I mean, your answer is right too, but mine just goes with the minimal effort possible
I interpret the relationship as the possibility of being parallel, orthogonal or neither as defined in C
Yes i understand that
As they're neither, making w=2u preserves that wrt t
They haven't mentioned what same relation implies
True, but that's a semantics problem. Not a mathematical one
Also going by that logic, w doesn't have to be parallel to u, we'll get a locus for that
True, as I said, just making it parallel is the fastest answer
So basically, @dusk summit if you're sure that relationship merely means being parallel, orthogonal or neither, my solution is the fastest.
If according to your teacher it implies preserving another property, then go with Rudy's to be sure.
But i don't think a question like that is to be graded for 5 marks
Once again you're assuming semantics
true
But looking at a problem also means understanding every aspect of it whether it be linguistic or psychological or mathematical
i got 3168.20
My advice is above. If "relationship" merely a condition, go with my solution. If it's something more specific, go with yours
With the information provided, I'd say the onus is more on a poorly formulated question
Let us focus first on helping the guy then we can go on about our arguement
I'm about to sleep, that's the issue 😅
no worries
Do not merely multiply t by a scalar @dusk summit
That will make it parallel
I think Rudy and I agree on this one
wait is this good then
Has your teacher explained abput what "relationship" means before?
That's multiplying t by 2, so no, it's wrong.
Wrong arithmetics probably
But you're going into a dead end
Any vector multiplied by a scalar (a number) not equal to 0 produces a parallel vector
so what values shouild i use for w
Read this
And the message after that one
And the previous ones
yeah relationship does mean parallel, orthogonal or neither
@dusk summit 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.
$\int -\frac{e^{-2x}}{e^{-x}+1}\ dx$
SELVATOR
!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
I can't solve it
which step?
U can say 1
Howd y9u get the first term?
By Factorise
No are y9u perhaps showing different questions here ?
Original one had e^-2x in the numerator, this one has e^x + e^-x
Look at last sentence
Oh sorry my bad 
@empty prairie
Use u sub here
It would be x/(e^x + e^-2x)²
Where'd the square come from now...
I mean have you tried simplifying the fraction by resolving the negative powers?
Is this your original question or is the question posted original?
.
U sub seems quite forward
1/(e^x + e^{-2x})
What would U be and V' be
u-sub ce nest pas integration par parties
Changement de variable
u = e^x
du = e^x dx donc du/u = dx
Idk how to use it
Ok so lets not to it then
$\frac{e^{-2x}}{e^{-x} + 1} = \frac{1}{e^{2x}} \cdot \frac{1 + e^x}{e^x}$
@empty prairie
???

$\frac{1 + e^x}{e^{3x}} = e^{-3x}(1 + e^x) = e^{-3x} + e^{-2x}$
@empty prairie
confusion
Any steps?
What this is literally just re-writing $a^{-b} = \frac{1}{a^b}$ and then reciprocating and multiplying
@empty prairie
+1 was in the denominator
$\frac{1}{e^{2x}} \cdot \frac{e^{x}}{1 + e^x}$
Hold on lemme get some pen and paper
SELVATOR
$\frac{1}{e^x + e^{2x}}$
SELVATOR
Oh shit mb lol
U mean add e^x + e^2x and minus it?
First of all
Which integral do you wanna do
@fierce lake
Can u not just write this as some hyperbolic function
@fierce lake @fierce rain
e^-x = t
-e^-xdx = dt
whats the question,
@stable crescent
@cedar adder 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.
anyone know fast multiplication algorithm for medium size number? Karasuba seems to be only good for big size one, so Karasuba is no go to me
why not just like, schoolbook multiplication
do word-wise multiplication with carry propogation
it works fine if you are not using too many bits
it is asymptotically slower than Karatsuba, but overhead matters a lot more for not-so-large numbers i reckon
how many digits?
36 digits
because I want something faster
by hand or by computer?
it's actually computer
by hand method is also welcome
well, for hand i would reccomend karatsubaing until 3-4 digits and the school multiplication
hmm...
for computer, i dont think 36 digit multiplication takes enough time to be non-negligible? 36 bits or digits? cause 36 digits is like 120 bits, and then tom-cook seems nice, or maybe even the fft one, idk ehich is faster tho, i would reccomend just trying them both out, tom-cook is a safer bet ig tho
aight thx
@timid bear Has your question been resolved?
Closed by @timid bear
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.
given the equacion x^2+y^2=a^2 we need to integrate the area using a closed line integral
What is the integrand?
x^2+y^2 =a^2
do you mean to say you want to find the area of a circle with integration?
Oh I see
Well what’s a parametric description of that region?
If you can do that, then you can find it’s area using a closed line integral
(Due to Greens theorem)
huh?
To give you an idea, it's a sphere with center (0,0) and radius r.
I’m not integrating it any other way, I’m doing it by a closed line integral. I’m not trying to avoid it
It seems like you assumed I’m trying to to it without a closed line integral
ummm ok so your way then i will shut up
so what should i do first
.
Find a parametric description of the boundary of ur region
That’s how u start when calculating a closed line integral
x=t^2.y= -+root(a^2-t^2)
How do we know when to switch sign?
There’s a better way to describe it
Hint: ||trigonometry||
hmmm sin^2t+cos^2(t)
asin^2t+acos^2(t)
Yes, I’m assuming you have ^2 around all of a*sint, right?
yes
sin^2,cos^2
Not quite
cos^2,sin^2?
x^2 + y^2 = a^2 described our boundary
cos,sin
acos(t).asin(t)
Yea!
Let us assume t varies between 0 and 2pi
Now we have a parametric description of the boundary of the region you want to find the area of
Denote ur region by $D$
Aslan
Then $\iint_D , dx dy = \text{Area}(D)$
Aslan
But by Greens theorem we then have that, say $\int_{\partial D} x, dy = \iint_D, dxdy = \text{Area}(D)$
Aslan
So if you can calculate the closed line integral $\int_{\partial D} x, dy$ then you’ve calculated the area of D, ur region.
Aslan
And now since you’ve got a parametric description of $\partial D$ then it’s straightforward
Aslan
Do you know how to proceed with the line integral?
Are you familiar with a line integral of the form $\int_{\gamma} P, dx + Q, dy$ ?
Aslan
I showed that for completeness, you only now have to compute a certain closed line integral.
so you are saying in order to do a line integral for the function y= f(x)
we need two functions for y and x in terms of variable t
and then apply Green's Theorem
?
What do u mean by this
parametrization as you called it
I’m assuming you know how to compute line integrals
Is this ur first time?
first time for what?
Computing a line integral, like this one
ohhh no you did great
Let’s start over and go back to ur problem about the area later when we’ve gone through how to compute line integrals
yes
Could u give me a definition or a short description of it?
Contrary to what the word may suggest, a curve can certainly be a straight line
An n-dimensional (real) curve, say gamma, is a continuous function from some interval I to R^n
well some function not in the form y=mx+b
Does this make sense?
So for example, if n = 2
Then $\gamma : [0,1] \to \mathbb{R}^2, , \gamma(t) = (t, t + 1)$ is a curve
Aslan
yeah i checked it
So suppose you were only given the graph y = x + 1
Then working backwards, gamma above is a curve that parametrizes the graph y = x + 1
Makes sense?
Then this description
Parametrized the curve!
Makes sense?
yes
So why do we care about parametrizing an equation or curve?
Well because line integrals are defined in terms of a parametric description of a given curve!
Let’s go over that definition
hmmm
ohh yes i get it
This is just a definition
yeah yeah gama prime is dx,dy
ok
That relates this notion of a line integral with the usual double integral
And that’s how you can calculate the area of a region using a closed line integral
one question what if the line is not closed does the theorem break
It breaks in the sense that we don’t know what the “region” means in that case
The way the theorem is setup, is that we start with a region
Then take its boundary
This must be a closed curve/line
So going backwards is undefined if the curve is not closed
I.e it doesn’t enclose a region
ok got it
This is basically a formula
U can now use this to calculate ur area from before
Ur gamma is (a cos t, a sin t) with 0 <= t <= 2pi
While P = 0, Q = x
ohh right
@next vapor Has your question been resolved?
@next vapor Has your question been resolved?
@next vapor Has your question been resolved?
Closed by @next vapor
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.
Let $G'$ be the group of real matrices of the form$\begin{bmatrix}
1 & x \
& 1
\end{bmatrix}$. Is the map $R^{+} \to G'$ that sends $x$ to this matrix an isomorphism?
shit, sorry
What a wonderful world !
So can I interpret the blank as an arbitrary constant
?
Let $G'$ be the group of real matrices of the form$\begin{bmatrix}
1 & x \
& 1
\end{bmatrix}$. Is the map $R^{+} \to G'$ that sends $x$ to this matrix an isomorphism?
What a wonderful world !
@marsh forum Has your question been resolved?
Closed by @marsh forum
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.
@urban copper Has your question been resolved?
Hi what's the question
@urban copper Has your question been resolved?
pls translate
well question 1 is analogous to solving ([\vec{v}]{B}=[\vec{v}]{B'}), or:
[\begin{bmatrix}1 & -2 & -2 \ 1 & 1 & 2 \ 0 & 1 & 3\end{bmatrix} \begin{bmatrix}x \ y \ z\end{bmatrix} = \begin{bmatrix}2 & 1 & -1 \ 2 & 2 & 1 \ 2 & 3 & 1\end{bmatrix} \begin{bmatrix}x \ y \ z\end{bmatrix}]
PajamaMamaLlama
yes but instead of x,y,z its a,b,c if you will because is coordinates, do not confuse with a vector, right?
also is this valid?
either you or me made a mistake mate
ig if your teacher specified but they're just letters 😅
yes that's just rearranged
(\begin{bmatrix}1 & -2 & -2 \ 1 & 1 & 2 \ 0 & 1 & 3\end{bmatrix} \begin{bmatrix}a \ b \ c\end{bmatrix} = \begin{bmatrix}2 & 1 & -1 \ 2 & 2 & 1 \ 2 & 3 & 1\end{bmatrix} \begin{bmatrix}a \ b \ c\end{bmatrix}\implies\left(\begin{bmatrix}1 & -2 & -2 \ 1 & 1 & 2 \ 0 & 1 & 3\end{bmatrix}-\begin{bmatrix}2 & 1 & -1 \ 2 & 2 & 1 \ 2 & 3 & 1\end{bmatrix}\right)\begin{bmatrix}a \ b \ c\end{bmatrix}=\begin{bmatrix}0 \ 0 \ 0\end{bmatrix})
we get different matrices mate
PajamaMamaLlama
either you or me made an oopsie
we do? 🤔
subtract
also we haven't covered change of basis matrices

🤔
, w nullspace {{-1,-3,-1},{-1,-1,1},{-2,-2,2}}
entry a_23 was incorrect
and indeed now notice R3=2R2
thus we expect the dimension of the nullspace to be 1
which is what was returned
i think it should be x,y,z, not a,b,c
or no?
i am already confused
well no, we have:
[
\begin{bmatrix}
-1 & -3 & -1 \
-1 & -1 & 1 \
-2 & -2 & 2 \
\end{bmatrix}\overset{G}{\longrightarrow}
\begin{bmatrix}
1 & 0 & -2 \
0 & 1 & 1 \
0 & 0 & 0 \
\end{bmatrix}
]
G?
gaussian elimination/row reduction
they're just letters, I could've chosen n,b,p and it wouldn't matter
when we got the nullspace what did we find
the meaning of the letters
PajamaMamaLlama
wdym? the letters don't inherently mean anything
when we got the nullspace what did we find
the vector (or family of vectors) (\mathbf{v}) such that ((B-B')\mathbf{v}=\mathbf{0})
PajamaMamaLlama
in other words we found all v such that (v)B = (v)B'
is a line that passes through the origin, infinitely many v such that (v)B = (v)B'
it appear to be so, yes
only trivial intersection
or no?
maybe i made a mistake
OD is the bisector of AOC,OE is the bisector of angle BOC and OD is perdecular to OE
prove A O B are collinear
@urban copper Has your question been resolved?
Someone else is already using this help channel. If you need help with a question, please open your own help channel/thread (see #❓how-to-get-help for instructions).
@urban copper Has your question been resolved?
but I distrust it
why would they have no intersection?
most likely I made a mistake
I went through the previous discussion and the method and i arrived at the same results doing the computations
the solution space to this matrix is a line passing by 0 and there is no intersection between the plane S and this line
except 0 of course
is strange, when we found the nullspace, would u have used a,b,c or x,y,z
either a,b,c or x,y,z are fine on my part, the important thing is that you know what these represent in the problem i.e. the coords in both base of a vector with the property we want
it’s just notation imo, pick the one that makes the most sense to you, or the one imposed to you
sure but when we were doing the equality vB = vB'
a,b,c made more sense
but the result of the nullspace is a vector
like multiples of a vector
is confusing, I got lost
yes, but when we found the nullspace we got a subspace
wait up you might be right
the a,b,c are coords, scaling factor in a linear combination of vector from either base A or base B
yes
so computing the linear combination in both base should gets you the actual vector generating the line of sol in R^3
ngl this is a bit wacko
oh OK
so the subspace we got are the coordinates or the vector?
XD
case and point when solving the system we found the set of coords for which vector expressed in each base have the same coords, but the coords are not the og vectors
yes!
its the coords
(a,b,c)=(2,-1,1)
this is why insisted on the a,b,c vs x,y,z thingy
it gets confusing
yeah sure, you can use x,y,z
but then you end up believing other thing that is not
what are the possible Vs?
,w 2{2,2,2}-{1,2,3}+{-1,1,1}
the coords that works are scalar multiple of (2,-1,1), then i think this means that doing the last computation here gives you the vector that spans the line of vector in the actual R^3 which have the same representation in both bases
all the possible vs is this <(2,3,2)> i guess
and it’s in S
crazy mate
a yeah what a tricky one
really good exercise though, it really checks if you understand what a basis is
by confusing you between the vectors of coords and the actuarial R^3 where the basis vector are
you mean that in a basis {x1,x2,x3}
ax1 + bx2 + cx3 = 0
if and only if (a,b,c)=(0,0,0)
or somrthing like that?
where x1,x2,x3 are vectors tho
idk when we solved the system we got out a vector of coords which was ‘disconected’ from the R^3 where the basis were
yeah
ima go get chips and a coke
yes mate, I appreciate the clarification
its related but not disconnected I think, like when we solved the nulspace we got all the scalar multiples of (a,b,c)
that choice of words is just my limited english
you are not from the states? oh
i’m from québec, it’s a french speaking province in canada
it’s funny because imo french and spanish are closer from each other than they are from english, so when you send your problems it’s close enough i can make a rough translation that works
no you did this time but it’s way back up
well I just used the inbuilt translation from the phone haha, didn't even checked if its correct translation
was ok imo
what about 2) though? are you still getting chips? can u maybe help me a bit more 🥺
well
(-3,1,5) = -2(2,1,1) + v2 + 3v3
(8,5,9) = 3(2,1,1) + 2v2 + 2v3
this is way more simplistic compared to the first one, I think
but lets see
,, \begin{cases} (-3,1,5) + 2(2,1,1) = v_2 + 3v_3 \ (8,5,9) - 3(2,1,1) = 2v_2 + 2v_3 \end{cases}
renato
,w {-3,1,5}+2{2,1,1}
,w {8,5,9}-3{2,1,1}
,, \begin{cases} (-3,1,5) + 2(2,1,1) = v_2 + 3v_3 \ (8,5,9) - 3(2,1,1) = 2v_2 + 2v_3 \end{cases} \ \begin{cases} (1,3,7) = v_2 + 3v_3 \ (2,2,6) = 2v_2 + 2v_3 \end{cases}
renato
(2,6,14) = 2v2 + 6v3
(2,2,6) = 2v2 + 2v3
(2,2,6)-(2,6,14) = 4v3
no I made a mistake
-4v3
(2,2,6)-(2,6,14) = -4v3
idea is good though
(1,0,1) = v2
(0,1,2) = v3
B = {(2,1,1); (1,0,1); (0,1,2)}
(-4,1,1) = a(2,1,1) + b(1,0,1) + c(0,1,2)
,, \begin{amatrix}{3} 2 & 1 & 0 & -4 \ 1 & 0 & 1 & 1 \ 1 & 1 & 2 & 1 \end{amatrix}
renato
,w rref {{2,1,0,-4},{1,0,1,1},{1,1,2,1}}
first exercise was way more difficult, this one is entirely mechanical
yep, I appreciate it mate
!rats










hopefully my exam contains exercises like first one or harder
I have prepared a lot, it would be a pity if it contains only mechanical exercises
ty for the help
.close
Closed by @urban copper
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 will the domain of gof be?
(-Infinity,4] or R?
Should be all real number
pretty sure it's $( -\infty, 4]$
What a wonderful world !
When you compose functions, the function's domain is at most the domain of the innner function
Is g(fx) right
Alright
Thankyou!
.close
Closed by @surreal shell
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.
- resolved (I think)
Closed by @mystic pilot
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 ended up with r= lambda but idk what to do now
i know
2pirh = 150
V(r,h) = pir^2h
then we have
Vr=lambdaSr
and
Vh=lambdaSh
did i do it right
.close
Closed by @sharp rampart
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.
The question: Generally if p is evidence for q, and q for r, then p is evidence for r. But this is not necessarily so. For, p might also be evidence for some 'cancelling event' c, which is evidence against r such that all-things-considered p is neither evidence for nor against r, i.e., such that P(p | r) = P(p). In that case, p's positive influence on r through q gets perfectly cancelled by p's negative influence on r through c. What I call 'premise 3' prevents there being such a c, which seems to make the transitivity inference valid, assuming we know that p is not evidence against r.
The intuitiveness of this piece of informal reasoning suggests that the following inference is valid:
- Premise 1: P(q | p) > P(q)
- Premise 2: P(r | q) > P(r)
- Premise 3: 'There is no such c.'
- Premise 4: P(p | r) ≥ P(r)
- Conclusion: P(r | p) > P(r)
The validity of this inference can be proven if P3 was P(r ∩ p | q) = P(r | q) and P(r ∩ p | ¬q) = P(r | ¬q), i.e., if the only way p influences r is through q. But that's a much stronger assumption then saying that there is no such c. In fact, it's so strong that it makes P4 superfluous. I am looking for a formalization of the informal thought expressed P3 that does not make any of the premises superfluous but still makes the inference valid. I am super stuck on this problem... Hence, I am very curious to hear your thoughts :)
I mean P1 is false right away
can you prove that?
take p = ¬q
oh P is premise
Hence, there's no inconsistency
It's an event.
It's supposed to be an event such that p influences r through c.
in such a way that this influence exactly cancels against p's influence on r through q.
does that help @thorn bay ?
@gusty valley Has your question been resolved?
why are you using premise 4?
would you want probability of p given r is greater than probability of r?
well otherwise outweighing propositions would also be counterexamples to the validity of the inference. That is, otherwise it wouldn't be the case that the only type of counterexamples to the validity of this inference would be cancelling propositions. More formally, the following type of proposition would also consistute a counterexample to the validity of the inference: a proposition o such that p is evidence for o, while o is strong evidence against r such that all-things-considered, p is actually evidence against r, i.e., such that P(r | p) < P(r).
I am interested in a formalisation of the informal thought expressed by premise 3 and not in something much strong that would obviously make the inference valid.
i think you mean P(R|P) < p(R)
well, what I want is a formalisation of the informal thought expressed by premise 3 such that one can validly infer the conclusion from premise 1-4.
yes sry
Would be curious to hear your thoughts :) @summer tapir
well, premise 4 is:P( r|p) >= P(r)
so the conclusion is basically P(r|p) is not = P(r)
yes
well so you eant P(r and p)=\= P(r)P(p)
yes
i.e., r and p are no independent.
right?
BTW: I am confident that this cannot be done without using premise 3. That is, without using a formalisation of the informal thought expressed by premise 3.
so you want a neat version of Premise 3
yes!
and once we have that, I don't think the proof will be too difficult
hm
P(r|p) = P(r|qp)*P(q|p) + P(r|-qp)*P(-q|p)
well
cant you just say P3 is P(r|p) =\= P(r)
haha
that would be easy right
but that is not at all a formalization of the informal thought expressed by premise 3
the issue is although p affects q and q affects r, the part of q that p affects is not neccessarily of part of q that affects r
Like I said, what I need is a formalization of the informal thought expressed by premise such that none of the premises become superfluous. This definition of premise 3 makes premise 1 and 2 superfluous
lemme give you an example of what i mean
ye, no I understand why the 'makes more probable' relation isn't transitive
P: it was raining
Q: garden is wet
R: you poured water on plants in garden
but I am looking for conditions under which it's transitive
can you tell me what your intuitive version of P3 would say in this case?
I have found very strong conditions under which it's transitive, namely P(r ∩ p | q) = P(r | q) and P(r ∩ p | ¬q) = P(r | ¬q), but I am looking for conditions that are much weaker.
well, the minimally weak condition is "it is transitive"
you literrally cant get weaker than that
so can you tell some place of how weak you want it?
well, idk this is a difficult one because the context suggests that p is actually evidence against r, while in my example I want p to be evidence for r.
but like we may stipulate that the plants are likely to be happy when the garden is wett
p: it was raining
q: garden is wet
r: plants are happy
On this set up, it suggests that it's raining makes it probable that the plants are happy
ok, but what would premise 3 be in this case? (intuitively speaking)
i.e., it was raining positively influences the probability of the plants being happy through making it probable that the garden is wet.
well, c would be some event such that c is made more probable by p, while c makes r less probable, i.e., p has a negative influence on r's probability through c.
A final condition c has to satisfy is that the positive influence p has on r's probability through q has to be exactly as big as the negative influence p has on r's probability through c such that all-things-consider p has neither a positive nor negative influence on r.
Well, we can find an intuitive example of that
well the issue is c can intersect with q
not r, q
sry
let's consider the example of the kids playing in the garden.
Suppose the kids like playing in the garden when the garden is wett. Then, it was raining makes probable that the kids are playing in the garden. However, the kids destroy the plant. Hence, the plants are less likely to be happy given that the kids play in the garden
Now, there might be all sorts of other things going on of course and it will be important to take that into account, but let's suppose for the moment that these are the only events influencing thing probability of the plant's being happy
here is the issue, how would you well define that
the kids play in the garden then would qualify as a proposition c only if the effect the event of it was raining has on the plants being happy through the event of making the garden wet has exactly the same effect but in the opposite direction as the effect of it was raining has on the being the plants' being happy through the event of the kids playing in the garden
that's exactly my question
i dont think there is a nice well definition of that?
imagine you have avenn diagram
how would you represent what you are saying on the venn diagram
let me think
@summer tapir
so P(r) = 0.5
P(p) = 0.5
well, you arent considering that c can intersect q?
P(p | r) = 0.5
doesn't matter for this example
you can let them intersect
also what would "c" be in this case?
well, the end result of there being a c is that P(p | r) = P(r)
that's impossible in what you have drawn
also the conclusion we want wouldn't follow here
but ye that's not the case here either, but that's in virtue of there being a c
alright sry
gtg
exactly, so even without a C, you cant get the conlcusion you want
also you mean P(r|p) = P(r), right?
The question: Generally if p is evidence for q, and q for r, then p is evidence for r. But this is not necessarily so. For, p might also be evidence for some 'cancelling event' c, which is evidence against r such that all-things-considered p is neither evidence for nor against r, i.e., such that P(p | r) = P(p). In that case, p's positive influence on r through q gets perfectly cancelled by p's negative influence on r through c. What I call 'premise 3' prevents there being such a c, which seems to make the transitivity inference valid, assuming we know that p is not evidence against r.
The intuitiveness of this piece of informal reasoning suggests that the following inference is valid:
- Premise 1: P(q | p) > P(q)
- Premise 2: P(r | q) > P(r)
- Premise 3: 'There is no such c.'
- Premise 4: P(p | r) ≥ P(r)
- Conclusion: P(r | p) > P(r)
The validity of this inference can be proven if P3 was P(r ∩ p | q) = P(r | q) and P(r ∩ p | ¬q) = P(r | ¬q), i.e., if the only way p influences r is through q. But that's a much stronger assumption then saying that there is no such c. In fact, it's so strong that it makes P4 superfluous. I am looking for a formalization of the informal thought expressed P3 that does not make any of the premises superfluous but still makes the inference valid. I am super stuck on this problem... Hence, I am very curious to hear your thoughts :)
premise 4 is not satisfied in your venn diagram!
do you think it's possible to make a venn diagram on which premise 4 is also satisfied? That would consistute a counterexample to the intuitive idea that the premises entail the conclusion
Do you understand why there can be no c is this case? c is defined such that P(r | p) = P(r)
<@&286206848099549185> any good with probability theory and logic around? We are stuck...
Would be very curious to hear your thoughts :) <3
You need to mak emore precise what you mean by the existence or non-existence of "c"
"c is defined such that P(r|p)=P(r)" doesn't make sense because c doesn't appear in those expressions
furthermore adding a premise of the non-existence of something can break your probability measure, once you define c (you haven't yet, as I said), you're much better off with something like P(c) = 0 than saying "c does not exist", unless you're really careful about the probability axioms
@gusty valley Has your question been resolved?
the question is to get precise on c
If I knew how exactly to define c
then my question would be solved
I give a bunch of conditions that the definition of c has to satisfy
what I mean is that there is no event satisfying all the conditions required for an event to qualify as being a c
Ah, I hear. Unfortunately I'm ill-equipped to answer your question.
I suspect it may not be a probability question at all, it might be a question about model theory
or modal logic
which are not my specialties
are you familiar with those
Does this informal reasoning (ignore everything except for the first paragraph) make sense to you? I think it's very intuive.
let me read it again
Yeah it make sense, emotionally. Don't ask me to prove it
Do you know about formal systems?
ok
haha. That's exactly the question though hahah
green makes blue more probable
and blue makes red more probable
but, indeed it's hard
so how would you make c in this situation?
well, c needs to be made more probable by p and make r less probable
well, that you can do almost always, just take c = p-r
unless r contains p
ye
@summer tapir have we resolved this?
so your premise 3 would need to be incredibly strong if you said there is no such c
yes
in a sense
wdym in a sense?
well, there is no c if it's false that P(r | p) = P(r). And ¬(P(r | p) = P(r)) is not a very strong condition.
that's what I mean by in a sense
well, the point is to find a formalization of the informal thought expressed by premise 3 such that premise 1 and premise 2 are not entirely useless
we want to make the transititivity inference from premise 1 and 2 to the conclusion valid
by adding some additional premises
if our additional premises make it such that premise 1 and 2 are superfluous
then we have achieved our aim
do you understand the informal reasoning that I outline here? Does it make sense to you, intuitively? @summer tapir
well, it kind of makes sense, but not something i would trust (intuitively)
As I said, I dont think it's probability you want, I think you're looking for a formal system of some kind
@gusty valley I can tell you the basics of what that means, but I'm far from an expert in them
and I cant promise you a better job of explaining than chatgpt or wikipedia
ye no I know a bit about formal systems and modal logic. I don't think it will help
thanks for offering @thorn bay
@gusty valley Has your question been resolved?
@gusty valley 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.
how do you do 8
I would expand it then take the derivative
The easier way is to write x^-2 as 1/x^2
well expanding it will just make it a power rule problem
i used the product rule and got (x^-2)(3ex^2)+(ex^3 + 3)(-2x)
Mb, I was trying to say the same thing as you.
happens
what's the derivative of x^-2 ?
still no
could be simplified cuz this looks ugly ngl
it's not wrong tho
i ended up getting 3e + (-2ex^3 - 6)/x^3
i dont know what to do from here tho im ps its wrong
I would accept this but idk about the one teaching you
@plain iron Has your question been resolved?
Are you familiar with rules of differentiation?
Yes
yea i know product and quotient
Then, just differentiate both sides using the rules you know and plug in x=2 at the end
@plain iron Has your question been resolved?
fun
If f(x) = 2g(x) + h(x) then f’(x) = 2 g’(x) + h’(x) then you subby in :>
subby I like that
- Constant multiple rule
2.simple differentiation of g and f
For q18, make sure you check your work properly
Only read below when you have tried it
|| If you have a constant, C, then D/Dx of C = 0. The indefinite integral would be the opposite resulting in an arbitrary constant, C||
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 explain to me how a (1/n)^3 appears
on the bottom
i dont know how the sin(1/n) >0
correlates to that
Because $\sin^3\qty(\frac1n)=\qty(\sin(\frac1n))^3$
;(
;(
!redir
This channel is only for on-topic discussion. Please take casual conversation to #discussion or #chill.
And please delete those 2 messages
Cluttering the thread
Please stop trolling/posting random stuff in here
hi
Closed by @nova pumice
Use .reopen if this was a mistake.
i think they left the server as soon as i spoke
yes
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.
Im supposed to find 2 of these that work and find out the 2 that don’t work. I also need help understanding why 2 of those wouldn’t work.
Cos( )=sq root of 3/2=tan( )
Tan( )=0=sin( )
Cos( )=-1=tan( )
Tan( )=undefined=sin( )
I tried plugging in cos(30) for the first one but I don’t think it would work for tan. And the other ones just confuse me, I’m not the best at trig
Hmm the ranges of sine cosine and tangent will probably come in handy here
Or maybe just for that last one
Im allowed to use radians or degrees for them but I prefer degrees unless it has to be radians
Im j really confused
I feel like all 4 don’t work
Because sine and cosine only have ranges between 1 and -1
Are we just trying to find values that fit in the blank spaces?
Yeah sorry
Alright
Well the second one is pretty easy to realize that it works
You’re just looking for the values of sine and tangent that just equal zero
What do I put for that in degrees? 180 and 0? Or are they both 0
Yeah I think they’re both 0
Unless we need to find all values that work
Or just one
Yeah if you look at a right triangle and take one angle of it, the tangent of that angle is the opposite side over the adjacent side
I can grab a picture if that helps
Well in terms of this stuff
How would I know how to do tan
Like for example the first one
How would I do tan for the square root of 3 over 2
If it doesn’t exist
Well it exists on the unit circle but like
Idk idk how I’d write tan if it’s not x (cos) or y (sin)
Hmm well
For the first one like I said ranges will help
Think about the range of cosine
It only outputs values between 1 and -1
So is there any way cosine can output square root of 3/2?
I’m talking about all the possible values you can get from cosine
It has a maximum and a minimum output
Wouldn’t. That just be every x on the unit circle
Oh
So cosine can’t go higher than 1 or lower than -1
Well I used 30 degrees
For that
But I just don’t get how tan would work
So I feel like that’s one out of the 2 that don’t work
No 30 works
Yeah sorry
Uh for tangent I don’t see why there shouldn’t be a value for that
Well what would it be cuz I’ve tried searching it up
And it gave me decimal
And it’s only supposed to be in degrees
Or radians
Yeah I’m not sure you’re gonna get a nice answer for that