#discrete-math
1 messages · Page 169 of 1
but we only want to talk about texas
i thought that, but texas is in the set of states
for ex
Well, yes texas is in the set of states but you see that statement becomes true even if it has nothing to do with texas
i think he wanted us to use quantification
yes
just y?
maybe ask someone else but I'm fairly sure you can't do ExEy
ur saying there exists a president and there exists a state such that a pres campaigned in that state and and lost that state
This is not equivalent to what 21 is saying
you'd need more relation symbols to say that something is Texas and that something else is Abraham Lincoln
oh my bad, you're right
they're not asking you to explicitly state that for texas, but rather form something that would be true if that's true
in which case that's fine
I didn't understand the question, sorry! @weary tiger
this mf owes me points
Yes, 21 is fine then
22 is fine as well, except maybe 22 is tricking you because some implies more than 1, but if it's not trying to trick you then it should be fine as well
I think it's safe to assume some means at least one but possibly more
so ExEy yeah
could someone help me out
this has me so confused
with modular expressions
wtf does like a^-1 mean
@valid wave pre-images?
for instance solving this
I know when we have like a congruent to x (mod y)
it would be a - x = k (y)
but im so lost when we have like a 3x like here
This refers to all such x such that 3x-17 is a multiple of 20. One such value is x=19, if I'm not tripping.
There is a systematic method to solve linear diophantine equations.
wait so its still the same idea of saying
3x - 17
is mod of etc
so then how would I show all the solutions
are you danish? @valid wave
You should look up linear diophantine equations and methods of determining their solutions.
It's a fancy name, nothing high level tbh.
At least for the linear case.
I'll dig up some resource.
a(a^-1)=1modn
it doesn't always exist
if you have x mod n or whatever, then x^-1 is such that x*x^-1 == x^-1*x == 1 mod n, and x^-1 is called the inverse
in this case 3 and 20 are relatively prime so it exists
2 == 3^-1 mod 5 bc 2*3 == 6 == 1 mod 5
so you could multiply both sides by the inverse of 3 mod 20
You must place in a row the 11 children of the family F and the 4 children of the family B. Knowing that the children of the family B always quarrel, you must separate them by at least one child of the family F.
How many alignments are possible?
I'd like someone to redirect me based on this chart
the column on the left is relatively understandable as for the 3 black things
first one is "Is the order important"
second is "Is there a repetition"
third one is "Do you need to choose a subset of the element amongst all the elements"
i'm kinda lost
got it
did some practive problems and got it down
but i got a case where theres no multiplicative inverse
so im assuming it would be imposiblr
yep, that's totally a thing
❤️
Wait maybe im dumb though
cause 3 of my questions
dont have inverses
the 6 , 8 and 8
dont have inverses
uhhh
i mean for a
x = 7 works
so you're doing something wrong, yeah
what are you doing
i don't understand
was following this format
to get a range of values
I thought that was the process with congruence multiplication
now im just confused 😦
oh, a0^-1
yeah i don't think ppl usually say sub
they say like a0 or a_0
tripped me up
no
you're doing something wrong
because i just showed you a solution for a
x = 7
but that doesnt follow the form of a0^-1
??
since we cant find an inverse for 6
wait ok go back go back go back
with z =27
what did you get for the solution to 3x == 17 (mod 20)
{19 + 20k: k e Z}
ok
just checking, i was worried you had a more fundamental misunderstanding but you don't
so if you have ax == b (mod c), and a, b and c all have a common factor d
then you can do like ax/d == b/d (mod c/d) and the solutions will be the same
more or less
im saying
if all my other solutions have ranges
how would I get a range for this question
when 6 doesnt have an multplicative inverse
well the new equation is still a modular thing
which will have a range when you solve it
so you can just multiply those solutions back again, sorta?
look
just trying it helps in this situation i think
6x == 15 (mod 27)
if we turn that into
2x == 5 (mod 9)
then
...?
nvm goti t
this is wrong
how so
x = a^-1 * b
but it's like a = 2, b = 5?
wait
wait
i'm a fool ignore me
yes
ok
hehe
ok so you have 7 + 9k solves the new equation
so it turns out that 7 + 27k solves the old equation
but!
14 + 27k also solves it
wait
no it doesn't
it's 16
right
that makes more sense
because 16 = 7 + 9
yes
ok so 7 + 27k solves the new equation, but 16 + 27k and 25 + 27k also solve it
do you sorta see that that works, even if not how
because if you keep adding 9 then it's just 34 which is just 7 mod 27
so as x == 7 mod 9, x == 7, 16 or 25 mod 27
jesus i've forgotten all my modular arithmetic lmao
slowly but surely i am becoming the meme of the mathematician who can think in 7 dimensions but can't do sums

A basket holds a set of balls. Each ball is red, green, or blue. How many balls must there be in the basket in order to guarantee that there are at least 5 balls of the same color? Make sure to explain your answer. is this just 15...?
nah, 13
why
4+4+4 is the max without 5 of the same colour
no youre british
therefore mine is the opinion which matters, colonial heathen
big difference
literally in england rn lmao
i cant read
A person’s birth date consists of the month, day, and year in which that person was born.The domain for a relation R is a set of people. No two people in the group have the same birth date.A person x is related to person y under the relation if x’s birth date is earlier than y’s birth date. Is this relation partial order? Is it strict order? Is it total order? Justify your answers to each
this is reflexive right
why it is reflexive?
person x is related to person y under the relation if x’s birth date is earlier than y’s birth date.
so xRx iff x's birthday date is earlier that x's birthday date...
nitezba
is this transitive, and if so how
i like to think of transitivity visually but i dont see it here
or does (1,4),(4,4),(4,1) make it transitive
It does seem to be transitive.
oof why if ac congruent bc modm then not necessarly a congruent b mod m
I don't understand why
c may not be invertible mod m
like if c is zero
ac=bc mod m always
c being "zero" just means c is a multiple of m
@weary tiger would you like a counterexample?
how can I solve 9^203 mod 10 without a calculator?
all I got so far is = 9^200 * 9^3
didn't get much easier at all
try looking at the last digit
whats the last digit of 9^1, 9^2,9^3...
ok
try find a pattern there
the last digits swaps between 1 and 9
What is 9 mod 10?
And do you know any property about multiplication and the modulo relation?
if you can answer those two questions you probably can solve your problem.
just testing some values and it seeming to swap between 1 and 9 is not a proof
yea 9 mod 10 is 9 but uh
-1 mod 10 is also 9
and (-1)^n is easy to compute...
@weary tiger interesting song
what song
in ur status
aw i was reading that
ah sorry lol, I found something that would help answering it. So I was going to try and solve it with that first. I'll post again if you want
I think I can use Pigeonhole Principle
yea i was thinking that
consider the parities of each co-ordinate
pidgeonhole and parity
no like is there some kind of intuition behind it?
I mean sure give me a counter example
@mental thicket
if two points have first point both odd/even, second point both odd/even, third point both odd/even then the midpoint is an integer
but there are only 2*2*2 choices 9 points so by php 2 have the same parity for each coordinate hence midpoint is an integer
yeah so a simple one would be 6*25 = 14*25 mod 100 but 6 ≠ 14 mod 100
the intuition is that you can have two nonzero numbers multiply to 0 mod m
What steps would I have to take to solve a problem like this?
there are no steps
well ok like
both problems are basically just combinatorics problems
so would it be 10C4 and then 10C2? 10C4 because each vertex is composed of 4 numbers? And then to be adjacent you need the intersection of A and B to equal 2?
OR instead of 10C2, 4C2? Because you are just looking at two of the same digits of a set of 4?
there are 10C4 vertices yes but there are way more than 10C2 edges
if anything there are 1/2 * (10C2)^3 edges
no wait
not (10C2)^3 what am i talking abut
10!/2!^3
aka 10C2 * 8C2 * 6C2
each edge is identified by three selections of two points: $A \cap B$, $A \setminus B$ and $B \setminus A$, the latter two (and only them) being interchangeable since we're dealing with an undirected graph --- hence the $1/2$ at the beginning.
Ann
hey i have a question
if there is a directed loop
would it be both a degree in and out
it should be less than m and greater than 0
well generally when you do something mod m, you get something greater or equal to 0 and less than m
because intuitively it's basically the remainder when you divide by m
yes true
so it's done
but here its saying that a.x mod m =1
no
because intuitively it's basically the remainder when you divide by m```
sorry I got confused
let me tell you my idea
and then you tell me what to do
so we said that the remainder of say a mod m= r , is less than m so r<m
and that's correct I understand that
so here we are saying a.x mod m= (a modm times x modm ) modm=1
sure
ok now how to continue
true
because anything mod m is greater or equal to 0, and less than m
yep true
now all the article says is it can't be 0 specifically
because a * x = 1, but anything * 0 = 0
ok so here x mod m is less than m
yes I understand
but we still haven't concluded that x <m now we concluded that x modm <m
oh
well we define x as being the smallest one
x = (x mod m)
we call that the inverse
ah
it makes everything simpler
uhhh sure
no

@short hamlet do you still need help w/ this?
Having this sets, how can I calculate the card of (A intersection Bcomplement intersection Ccomplement)?
nvm
How many possible sequences of heads and tails are there in k coin flips? How would i show this using bit shifts?
bit shifts?
like i guess you can shift numbers left or right by a certain number of bit
im very tempted to say 2^k
it is 2^k yea
coin has 2 sides so 50/50 chance
i guess you could consider a head as 1 and a tails as 0
thats true i guess the part that confuses me is why they ask "How many possible sequences" when there is no number of coin flips
ah ok ok thankyou!
How many bits do we need to represent the hex number 0xDEAD in binary?
would this be 16? i dont think this is correct because i am not including the "0x"
you want to represent the number 0xDEAD in binary not the string "0xDEAD" in binary
do you know what 0xDEAD means?
(and technically this depends HOW you represent it. i'm assuming you want to represent it as a base 2 unsigned integer)
So im going off of my notes which say, "Hex is base 16. There are 16 possible values for each hex digit
0,1,2,3,4,5,6,7,8,9,A,B,C,D,E,F. We can group 4 bits in a binary number into 1 hex digit." So its a shorter way to show binary
If i converted hex to binary would it be 110111010101101 (which is only 15)
if that is correct then yea, means you can with just 15 binary digits
0x is not part of the number. it is just used to say that what you are writing is a hexadecimal representation of a number
okok that makes more sense then. I was questioning if 0x was a part of it. I may need to double check on that binary just in case. I dont know for sure if it is correct
thank you!
yup 0x is for hexadecimal, and 0 is for octal
could anyone give me a hint as to how to derive that when a constant coefficient linear recurrence relation has a repeated root, you just multiply by n?
ex. $a_n=4a_{n-1}-4a_{n-2}\implies a_n=c_12^n+c_2n2^n$
nix
i know reduction order from diff eqs but not sure how i could apply that here
anyone in here good with sequences willing to get in a voice call and help me learn how to solve sequence problems efficiently? i cant figure them out half the time and the other half of the time i take way too long
if r is the repeated of the root of the characteristic equation then equating coefficients of t^2-At-B and (t-r)^2 gives you A=2r and B=-r^2, then you could just plug in nr^n and show it satisfies the relation
Sure. A verification is easy, but I feel unsatisfied in starting with the answer ahead of time. I'm looking for a derivation
With that RR if I suppose $a_n=v_nr^n$, then I get
$$v_n=2v_{n-1}-v_{n-2}$$
Which has, again, only one solution of the form $r^n$, being $v_n=1^n$ (which makes sense in context).
nix
But at least with all repeated roots for second order eqs, it can be reduced to this concrete one
See exponential generating functions
They convert homogeneous linear reccurences into differential equations
Quick Q
So we introduce x4
Where x4 can be integers 0-11
So can I solve the solution normally and then subtract ?
Why subtract
Oh to account for me making an aux. variable?
since the aux variable solution =/= the inequality solution?
The set of aux variable solution is same as inequality solutions
oh
Because that 4th variable is uniquely defined by the first 3 variables
So for every solution for original eqn,you get exactly one solution for aux equation
And vice versa
Ooh I kinda see, yeah its dependent. But then if you directly match the solutions, isn't it saying "three buckets may have less than or equal to 11 fish -> four buckets with 11 fish"
Is that the wrong analogy?
That analogy works fine
I don't see a problem with that
Nvm
Ah okay. shoot then I'm a little stuck?
You are mapping each case in "three buckets may have <=11 fish" to a case in "4 buckets combined have 11 fish"
Not directly
I want to find all the equivalence classes for this equivalence relation, but I'm not really sure how
<@&286206848099549185>
$x$ and $y$ belong to the same equivalence class if and only if $x^2 = y^2$, that is, $(x-y)(x+y) = 0$, that is, $x =y$ or $ - x = y$. So the equivalence class containing $x$, is ${x, -x}$. You can list all the equivalence classes as ${{x,-x } | x \in \mathbb{Z}_{\ge 0} }$
andreO
note also that for $x =0$, the equivalence class is just ${ 0}$
andreO
when I solve $(x-y)(x+y) = 0$ what I do is get 2 differents ecuation, $ x - y = 0$ and $x+y = 0$, so my answer is $x = y$ or $x = -y$
Jackieto
is that any different than your answer?
it's the same
alright was making sure, thank you
unclear what you mean by an=a^2n-1
this
Note $a_1 = 2^2$, $a_2 = a_1^2 = (2^2)^2 = 2^4$ and show by induction that $a_n = 2^{2^n}$
andreO
not entirely sure what you mean
I just want to know what the formula
You have the base cases. And now assume $a_n = 2^{2^n}$. Then from $a_{n+1} = a_n^2 = (2^{2^n})^2 = 2^{2^n\cdot 2} = 2^{2^{n+1}}$. So for all $n \in \mathbb{N}$, you have $a_n = 2^{2^n}$
andreO
jbr22
sorry im weak at this
okayy
hi everyone, do you know a good resource on discrete mathematics?
be it a book, or a course, i just want it to be easy to understand and explains every topic well.
You might want to check out MIT OCW's Discrete Mathematics course.
There's a complete set of video lectures available on their YouTube channel as well as their website. The website additionally has lecture notes and problem sets as well.
Ok i do not understand my professor's notes to do this when its simple. I cant wrap my head around it. 10 subscript 2 * 8
i can make 8 into 2^3?
@balmy hornet yes if i inderstand you correctly and 10_2 means binary number you can convert 8 to binary and multiply them or convert 10_2 to dec, multiply by 8 and convert to binary
there exists a bijection between the k-fold Cartesian product in A and all words in alphabet A of length k, correct?
yes
Great, was just checking, thanks
help
i understand this like 3 days ago when talking abut it in class and now i forgot alr bruh
what's that saying is that for all a,b there exist X
what does the next 1 says?
functiona=functionb
therefore a=b
??
depends on the value of a and b lol
for all and and b in X, f(a)=f(b) implies a=b
X being the domain of f
so assume 1/(a+1)=1/(b+1) with a and b in [0,inf) and show that does/does not imply a=b
like what are a and b? integers? rationals? even numbers?
Give the negation of:
a. all worms live in the ground
b. there is at least one pigeon that does not fly
any idea?
does anyone knows negation???
a) not all worms live in the ground (or there exists one worm that doesn’t live in the ground)
b) there is no pigeon that does not fly
i dont understand why it has to be nagative?
ca you explain me inverse and converse?
If the battery is charged then the flashlight works
a) give the modus ponens
b) give the modus tollens
i dont really get generating functions; if i am to find one that gives the ways to add up to a number k ( 3 + 2 + 2 is seperate from 2 + 3 + 2 ), how would i create a function?
i was thinking of saying something to add up to 1 and then multiply out but i am unsure if I can do that
im trying to figure out this problem, it says to factor each polynomial by finding the gcf. but i dont quite know what that means, do you factor the problem and then take that and find the gcf? the problem is 3b^5-6b^4+18b
the carrots are the squared bits
gcf = greatest common factor
@haughty garden any idea of where this logic of true and false come from?
I got a generating function of $$P _{k}(n) = \sum^{n} _{i=0} P _{k-1}(i)$$, I've gotten it added up until $P _{4}(n)$, but don't really get how to go higher. I also got a gamma function representation. Let's say I wanted to go for $P_{11}(n)$, how would I do that?
Hello, I did this proof
I am not sure if I need to add some words at the bottom
is this good enough?
To be honest, idk really how this proves anything but I know that getting the two sides equal is a good thing
how do u go from the first to the second
thats not the inductive hypothesis
I
Any help on this one?
Different strings using all the letters total is just n^r = 390625
so is it 390625 - only 1 A consecutively - only 2 consecutive As = our answer
sorry I meant cases where there is only an A isolated with no other consecutive As
yeah, but again, getting that is just a really annoying way of doing it
intuitively?
because it feels like you'll need the 2-ways and 3-ways to get that in itself, so it's not even worth it
just try thinking about the positions that the block of 3 can be in, directly
Ah gotcha, sorry Im used to looking at these problems via the subtraction method
So we have an 8 space string, to fit the 3 A character block
repetition is allowed but its a permutation problem because we have to use all the letters right? Can I just use the r-perm formula?
there are 6 ways to place the A's, and for each of those, the number of ways to place the remaining 5 letters is the same
so it really amounts to finding the number of unique ways to permute RRDVK
and then multiplying by 6
@naive gulch
ooh
thats an interesting way to approach it thanks!
That makes sense Botnuke <3
ty both let me try that
got that one, am now completely stumped here
what happens if you try to get the cartesian product of sets x sets within sets
so like A x B where A is {1,2} and B is {{0,2},{2,4}}
Part (a) or (b)? In either case, do you have an idea of which direction you want to prove?
ie do you have an idea of whether the statements are true or false?
Well in this case it's exactly what you'll expect, just all pairs of (a,b) such that a is in A and b is in B. So the cartesian product is just (1,{0,2}), (2,{2,4}), (2,{0,2}), (2,{2,4}). Just treat sets as elements too
hmmmmm okie ty
so a cartesian product always results in a bunch of () bracketed pairs
so wait if im going to disprove, would a counterexample suffice?
Yep!
but don't just try to blindly guess a counterexample. See if you can articulate why you think it should be false - what your hunch is. If it isn't, a good place to start is to unpack the definition of A - C and B - C; writing out what it actually means might give some insight
ok wait if i subtract a set with extra elements from a set that doesnt have some of those elements it doesnt matter right
like {1,2,3} - {3,42,69} = {1,2}
yes
Yep! this statement and your example are both correct
and you're definitely thinking in the right direction 😄
so ik part b is true but can it be done w just a direct proof
bc reading it out is intuitive enough but that's never enough ofc
Yep! part (b) is true. Start by writing out the definition of A - C and B - C. To show subset, remember you can just show that if x is in A - C, then x is in B - C '
do you have an explicit counterexample for part (a) yet?
ye i got that down w help from my gf too, it's basically when C has elements that are in A and B but individually if that makes sense
yep! that's exactly it
that's why I said you were on the right track here - you can take C = A U B, and then A - C and B - C are both empty, so A - C subseteq B - C, no matter what A and B are
so anyways, do this, and remember you are assuming that A subseteq B. The proof here is pretty fast
no problem!
also I'm not saying "it's a short proof" to make you feel dumb if you don't get it quickly
short doesn't mean easy
just konw that if you're doing 3 pages of difficult rearrangements or something, you're probably barking up the wrong tree. THere is a relatively succinct argument
succinct but feels incomplete
That is a little incomplete. In particular, you haven't shown the link between
"if x in A then x in B"
and
"if x in A - C then x in B - C"
your last two sentences are literally jsut the thing you are trying to prove
lol im tired
so wait if x is in A and x is in B, but is not in C, then the subtraction of C from either A or B will leave x in the respective set
Yep, exactly! you break into cases
or sorry, cases is the wrong word
I deleted that message since it was literally teh whole proof haha
so yeah, think about x being in A - C. Is x in A? what does that mean about whether or not x is in B? can x be in C if x is in A - C? with that information, what can you conclude about whether or not x is in B - C?
still feels a bit funky
you have the idea, I think the wording can be cleaned up
Talking about the set diference "removing" an element is a little imprecise
exactly my thoughts lol
Instead, literally just look at the definition of A - C and B - C
and see if x satisfies that definition (of being in B - C)
oh wait
if it does, you can just write "since x in B and x not in C, by definition x is in B - C"
im overthinking it
yes haha
no worries! I'm very glad this was helpful
for set A = {∅, {∅ , {∅ }}}, and set B = {∅ , {∅ }}
why does B ⊆ A = false while B ∈ A = true?
well ask yourself is every element of B in A
Hello, can anyone help with something?
Probably if you say with what
But now I've answered two questions. So you've even used up your bonus question quota for the day. Definitely no more help from me.
got room for 2 more questions? including this one
you got any idea what happened here?
specifically the inductive step
what is this notation
essentially what you do to construct the powerset of ${a_1, a_2, \dots, a_k}$ is you take every element of the powerset of ${a_1, a_2, \dots, a_{k-1}}$ once "as is" and once with the element $a_k$ added
Lochverstärker
You mean the notation P?
Could you please elaborate?
not sure how
try it yourself
write down the powerset of {a_1, a_2}
and use that to construct the powerset of {a_1, a_2, a_3}
try induction
oki
Can someone help me understand a generating function?
Go ahead
Ok so I have the function; $$ P_k(n) = \sum^{n}_{i=0} P_{k-1}(i) $$ and have a gamma function representation, but im confused as to getting P_11(n)
Ok so I have the function; $ P_k(n) = \sum^{n}{i=0} P{k-1}(i) $ and have a gamma function representation, but im confused as to getting $P_{11}(n)$
let me type it out nicer so that you can see it
Buncho Drunk
This?
$P_k(n)$ just looks like ${n+k-1 \choose n}$
Buncho Drunk
so would i prove that with induction?
oh yea, thanks
I have 4 12-sided dices
if they are all different, I can get 12 x 12 x 12 x 12 = 12^4 throws by intuition
uh but what is it even like
order matters so it can't be a combination
elements are distinct so it can be a permutation with repetitions
and since I can't encode the result in a set, it must be a n-permutation
however I don't know how I could get the same answer using nPr
help
@minor lake Can you you a bit more clear about what the question is?
if they are all different
how many results can I get
using something like a formula like nPr
with 4, 12 sided dices?
yes
Are you throwing them successively ie. one after the other
If so then the answer is 12x12x12x12
i know i just said it
If you are throwing the 4 dice at the same time then the answer is different
id just like to know if there is a way to use a formula or something,
im currently using a chart
it gives me the stuff to apply depending on the order, whether or not there are repetitions and if I have to choose a subset among all elements
and based on if it's yes or no it gives me a definition like n-permutation if there's order or combination if there isn't any
nvm i posted in #help-0 ill just c if someone finds something
solved
so if I have like a log function that is basically f(x)=logbase b(x-2) the domain would just be (2,infinity) and range would be all real numbers? range would also work as the target for the funtion of the logarithm right?
Also isn't a log inherently always going to be a one-to-one and onto function?
Yes,as long as b is valid
Yes inverse of f(x)=log_b(x-2) exists(g(x)=b^x+2)
So seeing as both domain and targets are infinite in scope, the size of the function can just be easily assumed to have |a|=|b|
but both sets of elements are infinite right? despite the domain starting after 2
Yes
weird
Hi guys, where would I go to get live help for discrete mathematics?
in an undirected graph, if there is an edge between A and B, does {A, B, A} count as a walk that is neither circuit nor cycle?
p = {(x, [x]) ∈ A × (A/R) : x ∈ A}
What does A/R mean in this?
is R referring to the real numbers?
I'm assuming R is an equivalence relation, and you're taking the quotient of A and R
the notation [x] usually refers to an equivalence class
And that's ust what makes the most sense
not one-to-one ,and another example that is one-to-one```
yeah that makes sense for the question. my professor very briefly talked about quotient space so I'm a bit confused still on what exactly A/R is
So the idea of an equivalence relation is that you're defining a new notion of "things being the same". That's why the equivalence relation axioms are the rules that a normal "=" sign follows. When you take the quotient of a set and an equivalence relation, you treat all things that are equivalent under that relation as literally being the same object. If you give me an example of an equivalence relation you've seen, I can give a specific example
so [x] is all the elements of the set that x is in, that relate to x? I'm probably confusing something lol
nope, that's exactly it! But we think of [x] being a single object, not the whole set of related objects
and when we say 'relate' to x, are we talking about equivalence or could it be something else
We're talking specifically under the equivalence relation
if [x] is a single object isn't x equivalent to [x]
So we treat [x] as a single object, but it is still the set of all things which are equivalent to x under R
In particular, [x] isn't any one element of A, it is a set of elements of A; the idea is that when you quotient A by R, you just treat all those elements as the same, so it doesn't matter which one you pick, they're all completely interchangable
okay I think the [x] part makes sense to me now.
The way my book is wording A/R it sounds like the set of all equivalence classes in A?
that's exactly it!
i can't find an example in my notes to make it a bit more tangible
because I think of something that wouldn't be one to one for p: A -> A/R
I can type up a quick example, give me one second
so p is a function that takes an element of A and spits out the set of everything equivalent to that element?
Yep, which we call the equivalence class of that element
does your course use the notation a ~ b for a being related to b under ~, or do you use like R(a, b)?
I just wanna type this up in a way that's consistent with what you've seen
for example, consider the equivalence relation R on the integers Z defined by aRb if a - b is even. We have that 2R2, since 2- 2 is even, that 2R6 because 2 - 6 is even, and in general, that if a and b are even, then aRb, since the difference of two even numbers is even.
Similarly, 3R1 since 3 -1 = 2 is even, that 3R3 since 3 - 3 = 0 is even, and in general, if c and d are odd numbers, then cRd
In this case, Z/R is just two equivalence classes; namely, [0] and [1]. The reason is that, since all even numbers are equivalent, [0] = {..., -4, -2, 0, 2, 4, ...} and since all odd numbers are equivalent,
[1] = {... -3, -1, 1, 3, ....}. Notice that [0] = [2] = [4] = [6] = ..., since the set of elements equivalent to 0 is the exact same as the set of elements equivalent to 2, since 2 and 0 are equivalent
What you've done is taken Z, and collapsed it down to the elements that are distinct under R. There are only really two distinct elements - 0 and 1 - since everything is equivalent to either 0 or 1, so in some sense, under R, there are only two elements. Notice that my choice of 0 and 1 isn't special here, I could take 2 and 3, or 4 and 13, or 1283781232 and 1111111111111111. What matters is that everything is either in the equivalence class of even numbers, or the equivalence class of odd numbers, and I'm just denoting those sets in the easiest way possible.
ah okay yes thats a good example thank you
The intuition that I always keep in mind is that A/R is taking A, and then making things that are equivalent under R actually the same object - I know I've said this, but it's a subtle idea, and once you have it this'll become much clearer
Hiya guys, in not sure if I'm using predicate logic correctly. I'm trying to make sure the notation and the usage of contradiction is proper.
Is anyone in here good with combinatorics?
I have a problem where i have to combine legos to form cuboids, where no repetion is allowed, what would be the best way? Permutation or Combination formula? But there is a problem though, only cuboids are valid so random mixes like figure (b) in the picture below is not valid, how would i account for that with either of these formulas? Also figure (c) and figure (d) are valid but the same so they are a repetition i guess
hi
induction
it's similar to the one i asked yesterday
but i can seem to even get started on this one
start by verifying it for n = 1
then write down the sum up to n in terms of the sum up to n-1
Can anyone help me find the symmetric closure of a matrix?
Trivial by intuition
is there a place here to discuss theory of computation?
You have two lists A and B.
A is an unsorted list, whereas, B is sorted.
Given ′𝑋′ a number input by the user, you need to determine whether or not 𝑋 is present in A and B.
In how many ways can I do this?
#foundations might work, in case you don't get an answer here @opal cedar .
many ways
well, at least two i guess
if B is sorted you can look for X in it through binary search
I don't have the "Advanced" role, so I guess I'll try here. I'm trying to convert a CFG to a CNF/GNF CFG then turn it into a DPDA, but I struggle with ensuring it's deterministic.
many ways
@stray reef
And another way for B?
And is there only one way for A?
Which one
well you could just search through B the same as A
for A you have no option but to just scan through it all
thank you, I appreciate it
Which way is the best according to you
why according to me
for A, simply going through it element by element is the most efficient you can possibly imagine
you can make it WORSE on purpose if you really want to
And for B?
for B you could do binary search, or you could do radix search (idk if that's the right name for it, but it's the formal equivalent of looking up a word in a dictionary by matching the first letter, then the second, etc.)
idk why you're asking "how many" ways there are to do it tbh that's kind of a weird question to ask
you could make a plethora of formally different algorithms for searching in a sorted list
are you SURE you're being asked "how many possible algorithms are there to search for x in B"?
or is this another mistranslation issue that i'm running up against and which you'll be unable to clarify because of an invisible language barrier
You can do this in more than one way
so you were not asked to count all the possible ways
idk what your teacher means by "work only for A"
an algorithm that works on any list whatsoever will also work on B
i can't think of an algorithm that only FAILS on fully sorted lists and works in all other cases
Nobody seems interested in my PDA question 😆
Note that $$\frac{1}{(2i-1)(2i+1)} = \frac{1}{2}\left(\frac{1}{2i-1} - \frac{1}{2i+1} \right)$$ Now you have a telescoping sum where consecutive terms cancel out except the first and the last and you're left with $\sum_{i=1}^n \frac{1}{(2i-1)(2i+1)} = \frac{1}{2}\left(1 - \frac{1}{(2n+1)} \right) = \frac{n}{2n+1}$
andreO
Isn't that Dirac
Someone asked a question in #foundations if you didn't notice.
How do we prove Euler’s Theorem holds if G is not connected?
is zermelo-fraklin set theory the foundation of mathamatics?
what do you mean by foundation?
I can explain this problem to anyone who wishes to help
anyone know much about discrete geometry/computational geometry here? I'm trying to pick a topic for undergrad thesis. Anyway, i had a simple question; if i had a simple polygon and shoot a ray from a known vertex to another known vertex, whats the fastest (order) algorithm for finding out which edge it first intersects, if it does intersect one?
A, do a linear scan, B do a binary search. If it is possible to have A sorted then do that and apply binary search to both 😄
Suppose I have the linear congruence 5x [congruent] 8 (mod 5) can't I multiply both sides by 5 and then I get 25x [congruent] 40 (mod 5) and since 25 [congruent] 40 (mod 5) cause 5 divides (25-40) then I obtain x [congruent] 1 (mod 5) Why can't I do this ? what's wrong with it ? I don't understand
well
in general suppose you have ax = b mod n congruence
provided a and n are relatively prime
then there exists inverse of a modulo n, a^(-1)
if I have a=b mod m and c=d mod m => ac=bd mod m
is the converse true?
suppose a and d not equal 0
a = d, b = c
so you multiply the whole equation by this inverse and get
a^(-1)ax = a^(-1)b mod n
or in other words x = a^(-1)b mod n
I’m...
Krieg your example only works mod 1.
Nvm. Poster asked about the converse. My bad.
mod 1 
Every equation always holds mod 1
I am currently studying Discrete Math by the book, this is in Fundamentals chapter. I am quite confused about this question.
this isn't a question, unless the actual question said to prove this
what are f_A and f_B?
indicator functions of A and of B respectively?
Hi everyone, I need to help about automata theory. I want to CFG rule tool. ( Ex: L(G)={b^n a^m b^n | n>=0, m>=0 ... As an answer S->bSb | A A-> aa|ε}
is the inverse of a transitive closure always equal itself? For a matrix
Still need help?
Could I get help on this problem?
pls
Hi
How can I prove scalability on modulus arithmetic relations?
i.e. a = b (mod m)
then ac = bc (mod m)?
c being a constant
Well, you mean that c is an integer, yea? Just use the definition of mod
I did...
this is what I got: m | c(a-b)
what do I do now?
I tried then using the definition of "divides by"
a = b (mod m) means that a-b = km for some integer k. So, now:
ac-bc = (a-b)c = kmc
Since c is an integer, m is an integer and k is an integer. Then, now, ac-bc = (kc)m and since kc is an integer, you are done
woah, hold up
...
Dude, my teacher sucks
I was never taught this definition and I kept seeing it in wiki pages
MY TEACHER SAID modulus is
m | (a-b)...
wait...
if I have mk = ac - bc
how did you get c over to mk?
that doesn't make sense
What do you mean? We have a-b = km for some integer k, yes? Then:
ac-bc = (a-b)c = kmc
I'm just using a simple fact about a-b which I established based on the conditions given
No
You are to prove ac = bc (mod m)
m is already in relation to ac and bc
you can't just add c to m...
Yes, we are to prove that ac-bc is divisible by m
so, we write an expression for ac-bc
not mck = c(a-b)...
No. Let $a = b (mod(m))$. Then, $a-b = km$ for some integer $k$. Now, consider $ac-bc$, where $c$ is an integer. Then:
$$ac-bc = (a-b)c = kmc = (kc)m$$
which follows from the regular rules of integer arithmetic. Since $kc$ is an integer, it follows that $ac = bc (mod(m))$.
Abhijeet
You just need to use the definition of divisibility
Your teacher said that a = b mod(m) is defined as m divides a-b. This is the same thing as saying that a-b = km for some integer k
obviously
Why?
because look
giving the definition of modulus, and of divisibility
you can only derive the following
bc = ac (mod m)
bc and ac are values!
so
m | (ac - bc)!
using modulus definition
using divisibility definition
we get
mk = c(a-b) after factoring!
you do not get mck
I have no idea why you keep stating that!
You just said that we have to show that if $a = b mod(m)$, then $ac = bc mod(m)$. Is that what you want to prove?
Abhijeet
urh bad latex but the message is clear
Yes
Then, what I have written above is correct
given then
Read it again
Bro
No, look, what part of it do you not understand?
Like, where is the flaw in the reasoning?
well... first off, it seems incredulous then
to have a statement
that goes
$a = b mod(m)$, then $ac = bc mod(m)$
MaggyD
MaggyD
But that's trivially true whenever c != 0 .
because all of our definitions revolve around m being the modulus applied to the values
so its absolutely stupid, FUCKING STUPID
I HATE THIS SHIT
I mean
math doesn't operate on the basis of what you hate or don't hate
The proof of your statement has been given above so have a look and tell me if there's any significant flaw that you find
ok
fine I can prove it lmao
your screwed
let's define
ac and bc as ac = q and bc = z
q = z (mod m) cannot equal a = b (mod m)
there is no way
Bro, the only way your proof works is if you do magic
you are adding c to M!
WHY WOULD YOU MAKE A PRODUCT OUT OF YOUR MODULUS
No one is saying that q = z (mod m) is equal to a = b (mod m)
Like the intuition makes no sense bro
I did not make a product. I'm just using definitions here
Homie
no your not
your addign c to m out of thin air
Prove me then, how my definition
given a = b (mod M) then ac = bc (mod m) that
You are wasting time
m | (a-b) = m | c(a-b)
There is no addition of c that is going on in the argument
it is not POSSIBLE
Period...
case in point
mk = ac - bc does not equal mk = a - b
THAT IS IMPOSSIBLE
You cannot be making this point right now dude,
No, you are wasting time. Either look at the proof and be specific about whatever flaw you think is present in it or leave the channel and do something more productive.
$given a = b mod(m)$, then $ac = bc mod(m)$
MaggyD
prove that?
That's not what's being said
You are given that a = b (mod m). Then, you need to show that ac = bc (mod m), where a,b,c are integers and m is a nonzero integer.
There's no sense of 'equality' here.
Kaisheng, the proof is above. I've already typed it out and I've been explicit. But this person hasn't looked at it/doesn't want to look at it/isn't telling me what's wrong.
What do you mean? We aren't applying c to m at all, whatever that means
No. Let $a = b (mod(m))$. Then, $a-b = km$ for some integer $k$. Now, consider $ac-bc$, where $c$ is an integer. Then:
$$ac-bc = (a-b)c = kmc = (kc)m$$
which follows from the regular rules of integer arithmetic. Since $kc$ is an integer, it follows that $ac = bc (mod(m))$.
MaggyD
Straight from the horses mouth.
stop me if you don't agree
You are incredulous
ok so slow down maggy
Seriously, it says it right there in your argument! You developed a product out of c and M!
lemme try pls
-.-
God only knows... it was his proof....
yes
ok
where doe she get specifically c on the rhs
so given that you have a - b - km
or as a product with m
if you have a - b = km
yes
ok
following
then you can multiply both sides by c and it's still true
so ac - bc = kmc
are you ok with this
sure
Ok... but this is not proof
why not
First off
look at our corollary statement...
ac = bc (mod) m
when we drive the equation from here... we get
mk = c(a-b)
No
which is not what you get in your proof
You are supposed to SHOW that ac = bc (mod m)
...
In other words, you are supposed to show that ac-bc = mk', for some integer k'
what
Where does it say that?
no everyone stop
from the fact that a = b (mod(m))
Dude it does not say that...
ok maggy lemme go through the whole proof
stop
2. a = b + km
3. a - b = km
4. ac - bc = ckm
5. ac - bc == 0 mod m
6. ac == bc mod m```
are you okay with this
no
ok
I give up
Can we do another proof that is similar
Frankly, I'm no fool, I understand the idea
ok then why do you not like the proof
given ac = bc from a = b both (mod m)
.......
tell me what you don't like about the proof
which step
2. a = b + km
3. a - b = km
4. ac - bc = ckm
5. ac - bc == 0 mod m
6. ac == bc mod m```
4,
from 4 to 5?
3-4
ok
so if a - b = km, then it is true that ac - bc = ckm
c is being applied to a and c in the propositions
this is a true statement
never is m modified
this is multiplication
homie... I know
ok, then what's the problem
but look at the expression
ac - bc?
please
sorry nitezba come back later
No. Let $a = b (mod(m))$. Then, $a-b = km$ for some integer $k$. Now, consider $ac-bc$, where $c$ is an integer$
MaggyD
Compile Error! Click the
reaction for more information.
(You may edit your message to recompile.)
maggy, i don't care about the initial proposition
you asked about the step between lines 3 and 4
welp that's what it says homei
you asked about the steps between lines 3 and 4
do you have an issue there
and I've spent an hour on this shit
do you have an issue there
I have an issue with the whole statement and how you could possibly apply a constant C to your modulus...
it is fucking stupid
at what point in my proof do i apply c to the modulus
...
2. a = b + km
3. a - b = km
4. ac - bc = ckm
5. ac - bc == 0 mod m
6. ac == bc mod m```
tell me
step 4
no
nit wit
yes there is
i did not write 'mod m'
