/sqt/ Stupid Questions Thread

Alternatively:

x^2 - x = 20
x(x-1) = 20

What number times one less than itself equals 20?

>you can't multiply any matrix by a 1x1 matrix

Sure you can:

A [a] = aA

>capstone project
>not thesis
brainlet

>Why is LSD even illegal?

"I think it's interesting the two drugs that are legal, alcohol and cigarettes, two drugs that do absolutely nothing for you at all, are legal, and the drugs that might open your mind up to realize how badly you're being fucked every day of your life? Those drugs are against the law. He-heh, coincidence? See, I'm glad mushrooms are against the law, 'cause I took 'em one time, and you know what happened to me? I laid in a field of green grass for four hours going, "My God! I love everything." Yeah, Now, if that isn't a hazard to our country... how are we gonna justify arms dealing if we know we're all one?!"

-Bill Hicks

Whats an example of a left-distributive operation (with regards to another) that is not right-distributive?

matrix multiplication

With regards to addition you mean? Thats both left and right distributive

an even more fun example is this, a simple projection operator onto the vector x,

[eqn]P=\frac{xx^\top}{x^\top x}[/eqn]

Go ahead and check that it satisfies
[eqn]P^2=P[/eqn]

Consider maps from R to R. Let the two operations be * = pullback or function composition and × = usual multiplication.

My school checks off most of that list.
When you say "it's shit" do you mean it's shit intellectually or does it seriously hurt job prospects?
I just wanna take care of my mom ffs

user, subtract 20 and factor using a +4 and -5

Say i have a heat distribution within a sphere modelled as u(r,t). Does "the temperature distribution in the centre is continuous" simply mean that u(0,t)_r = 0?
(_r means differentiate with respect to r)

I understand how to solve exponents when you multiply/divide bases but how the hell do you solve it when they are added or subtracted without using logs? Thank you in advance

>I understand how to solve exponents when you multiply/divide bases but how the hell do you solve it when they are added or subtracted without using logs?
Why wouldn't you use logs?

Logs aren't taught till the next grade so I'm assuming there's a way to solve this without it.

T. brainlet trying to relearn math 10 years after high school

>Logs aren't taught till the next grade so I'm assuming there's a way to solve this without it.
There isn't.

This was on an old grade 10 test. I guess the teacher is retarded then?

>There isn't.
Proof?

You can solve by intuitive analysis.
Play with it a little.

In the quadratic 12 - x - 6x^2, the terms are going to be factors of 12 (1, 2, 3, 4, 6, 12) on the left and 6 (1, 2, 3, 6) on the right. We can disregard the even factor options of 12 (2, 6) on the left because the x coefficient (-1) is odd. After this, I have to mentally shuffle round the right factors and order or signs till I find the right factors.

Question: after common-sense considerations like the above, how can I find the solutions apart from basically trial and error?

Use the quadratic formula?

>"""solve""" by ""intuitive"" analysis
That's not what "solve" means.

Is there a way to do what I described without using the formula?

-64=2^(x+2)-2(x+3)
-64=2^x*2^2-2^x*2^3
-64=2^x(2^2-2^3)
-64=2^x(-4)
-64/-4=2^x
16=2^x
x=4

He/she said without using logs

Trial and error then. Or he could go fucking retard mode and solve it by bissection.

If you are certain that the roots are rational numbers, you can assume that the quadratic can be factored as (ax + b)(cx + d) where a b c d are integers so that bd = 12, ac = -6, ad + bc = -1, from this third equation, you can see that b and d are coprime and similarly a and c are coprime. This can further reduce the search cases. You could also try reducing the equation modulo some number or use the intermediate value theorem to estimate the location of the roots. I don't think there's any general algorithm to find roots in this manner.

What is the abscence of abscence? If I can't answer this, they won't let me cross the loval bridge.

Yeah just factor out a 2^x from the right side,

[eqn]-64 = 2^x * 2^2 - 2^x 2^3[/eqn]
[eqn]-64 = 2^x ( 4-8)[/eqn]
[eqn]-64 = 2^x * (-4)[/eqn]
[eqn]16 = 2^x[/eqn]
now rewrite 16 as a power of 2,
[eqn]2^4 = 2^x[/eqn]
clearly x=4

also, get rekt kids:

>clearly x=4
He/she said without using logs.

prove [math]\sum_{n=1}^\infty \frac{sin(nt)}{n} =
\frac{\pi - t}{2}.[/math]
This came up on the 5th equality here:sos440.tistory.com/141

Somebody already has a solution on stackexchange (math.stackexchange.com/questions/566856/is-sum-n-1-infty-frac-sinnxn-continuous) but I don't follow (is that fourier transform on the first equality? I don't follow)

>Show that
[eqn]\frac{-1 + i \sqrt{3}}{2}[/eqn]
>is a cube root of 1
So I just computed it and showed the steps, is that what this is asking for? I feel like it's asking for some kind of property of complex numbers as an answer.

It's a meme... don't respond to this bait

Scalar multiplication and matrix multiplication are two very different operations. What you are doing is scalar multiplication with a 1x1 matrix and calling it matrix multiplication.

[math]1 = e^{2\pi i}[/math]
[math]1^\frac{1}{3} = e^\frac{2 \pi i}{3}[/math]
[math]e^\frac{2 \pi i}{3} = cos(\frac{2 \pi}{3}) + i sin(\frac{2 \pi}{3}) = \frac{-1}{2} + i\frac{\sqrt(3)}{2}[/math]

Let me know if any of those steps didn't make sense.

At what point should my brain automatically switch to those identities, or am I already fucked?

Well the i in the expression is a pretty big giveaway that you might want to use complex analysis to show it. So my answer to your first question is "as soon as you see an i in the expression".

As for whether you're fucked or not, just do more of these problems and you won't be.

>So I just computed it and showed the steps, is that what this is asking for?
Yes.

>>Scalar multiplication and matrix multiplication are two very different operations. What you are doing is scalar multiplication with a 1x1 matrix and calling it matrix multiplication.

I didn't say it was matrix multiplication. The point is just that there isn't much of a difference between 1x1 matrices and scalars; abusing a little notation never hurt anyone.

Matrices requires a choice of basis (at least in the context of vector spaces), scalars exist irregardless of any basis being chosen

>there isn't much of a difference between 1x1 matrices and scalars

Don't worry user, you will have your friends to support you!

>abusing a little notation never hurt anyone.
That's not what "abusing notation" means, retard.

Can anyone tell me just how much mathematics is used within Chemistry IRL? My major requires a Calculus I class, but will we actually use this IRL within labs or is it just to 'weed out' those who can't critically think?

How do I prove that the composition of two nonparallel Lorentz boosts isn't a Lorentz boost?

Is the cardinality of the set of all neighborhoods of a point greater than continuum in the standard metric space on R (i.e. d(x,y) = abs(x-y) )?

Science is pretty much all differential equations, so yeah you certainly need calculus.

Why is the current the same at the top and bottom of this circuit? Would it be split at the node after R1?

because of the grounding

because vout doesn't go anywhere

V_out is a node used to read/sense the voltage at that point. This structure is a Voltage Divider, as using precise values for the resistances will allow you to set a precise value of V_out in terms of V_in

Ideally, any device that senses/reads the voltage at a point (ref. to ground) does not itself draw current, so no current would flow out of the line labeled V_out.

contradiction

Sorry, I still don't get it. Thanks for trying, though.

Vout leads to nowhere, it cannot do any work there and it is an open circuit
so the whole current goes to ground

...

If by neighborhood you mean any set containing an open set around the point, then yes. Just form the union of each subset of R with an interval around the point.

> Just form the union of each subset of R with an interval around the point.
Uhh yeah, that's how you prove it by definition. The problem is I tried to use the Schröder–Bernstein theorem, it's easy to find an injection N(a) -> P(R) (N(a) being the set in question associated with the point 'a') , but I can't figure out what is P(R)->N(a) supposed to be. All the subsets of R in N are open sets with the cardinality of the continuum, what am I supposed to do with the finite and countable subsets of R?

Where can i learn about Lebesgue integrals? Khan academy doesn't have anything on them

The same reason why birds don't die when standing on high-power lines. In order for electrons (current) to flow, there must be a voltage difference to "motivate" them. That V_out wire is entirely on the same voltage, so there is no voltage difference.

>All the subsets of R in N are open sets
So you're restricting the neighborhoods to be open sets? If so then the answer to you question is no. Because there are only 2^N open sets.

>2^N
I mean 2 ^ \aleph_0 of course

There are some introductory lectures here
youtube.com/playlist?list=PLPH7f_7ZlzxQVx5jRjbfRGEzWY_upS5K6

What if I don't?

If you don't, then you can find an injection from P(R - I) to N(a) where I is an interval around a by mapping the set A to A \cup I

I think I got it. Thanks.

>C++
Is this the 20th century?

Any HLPL other than C/C++ is either babby coding or esoteric shit closer to math. Studying it as the first programming language gives you a good foundation. Nothing is too difficult after C++, except for the aforementioned esoteric shit like Lisp.

C and C++ are low level programming languages nowadays, and as such unsuitable if you want to learn high level programming (which is 99% of the programming one will ever do unless they are in a specific subfield like performance optimization or microcontroller programming, but that's too specific for CS). And for learning low-level programming, there's no reason why to use the bloated monstrosity that C++ is when C exists

Most software is still written in C/C++. C# and Python are pretty popular but still not on the same level. And Java... Well, frankly I don't remember the last time I used a Java program.

is that way.

>C and C++ are low level programming languages nowadays

Stop being retarded. Low level programming is AMD64, ARM, MIPS, PPC

amazon.com/Integral-Measure-Derivative-Approach-Mathematics/dp/0486635198

Good examples to take a video of Newton's First Law?

I tried to download all the protein coding genes in the human genome from ensembl biomart, unspliced transcript form
it's 874 mb of compressed fasta and still counting
1: am I retarded?
2: will the unspliced transcript form return multiple copies of each gene but different transcripts?

Try this.

I don't know if it's because I'm a brainlet or there is a massive grade inflation problem but at the end of the semester I'm always embarrassed to come to the grade threads because the best I can do is Bs and 1 A maybe. How do I get all A's next semester in EE?

[math]1 + 1[/math]

[math]
1+1 = 1 \cup \{1\} = \{\emptyset\} \cup \{\{\emptyset\}\} = \{\emptyset, \{\emptyset\}\}=2
[/math]

It's the last night before the exam.
How do I make myself stop feverishly cramming and just accept that I will have to go through the embarrassment of handing in an empty sheet of paper and try again in a couple of weeks?

You could cheat and memorize all the answers in your brain ;^)

Is it possible to somehow coat neodymium magnet with titanium nitride?

Let X be a random vector from (omega, A, P) to (R^d, Bor(R^d)) and let C be its covariance matrix
How do i prove that the following result is true for all t in R^d ?

E ( < X-EX, t>^2 ) = < CX,t>

< , > is the dot product.

>night before the exam
>early January
If you were autistic enough to take winter classes then you deserve it.

I could, but not in the next 5 hours. There are 80-something questions, and most require at least 30 mins to understand the answer. I can answer a few, but I can't memorize the proof of Gödel's incompleteness theorems for shit. It requires a lot of reading I don't have time for.

> everyone on Veeky Forums is an American, and the higher education system in every country in the world is exactly the same

Did you copy that correctly?
I think you must mean for the right side of the equation for things to have any chance of handling scalar multiples of t.

yeah sorry, that's

Then it's pretty straightforward if you use matrix-vector notation and expand things using linearity of expectation..

The critical point is that
^2 is t'(X - EX)'(X-EX)t
where t' is transpose(t).
Then use the fact that, for a random matrix A, and constant vector b, E(b'Ab) = b'(EA)b.

>living in a shithole country

Well, it's not like I got to choose.

If you just care about money it's going to be painful and embarrassing for you and you're never going to do well at it.

Your best prospect is finding a sinecure and hoping that no one really cares about how much you fuck up or how little you actually do.

Just bought Swakowski's Calculus with Analytic Geometry, 2nd revised ed. from Goodwill, how good of a text is it?

Usually I buy other mathematics books that I come across so I have larger problem sets to choose from whenever I need to go back over a topic and not forget it, just in case the answers for problems I already went over are still stuck in my head.

>Just bought Swakowski's Calculus with Analytic Geometry, 2nd revised ed. from Goodwill, how good of a text is it?
Why don't you read it and find out?

I'm about to. I was mainly looking for an overview of quality compared to some more modern texts or maybe a range from Stewarts to Spivaks perhaps.

I'll let you all know how it goes, though and what I think compared to the dozen or so other books I've gone through.

can anyone help me prove this? its a big-Omega question, I know it can be more CS but Im having trouble with the math parts.

I know that big-omega states that
f(n) >= C * g(n) for all n >= k
such that C and k are constants >= 1

for any two random values of k or C I choose the equation does not work, like if I chose
C = 1
and k = 5

log5(n) >= ( C ) * ( log2(n) ) for all n >= 5
and lets say n = 5:
log5(5) >= ( C ) * ( log2(5) ) for all n >= 5
1 >= 2.321 which is false

and for any values it comes out as false, is there a more formal way to derive C or K for this? thanks