#discrete-math
1 messages · Page 228 of 1
Friend and I have struggled on this one for a bit yet it seems so simple: show there are 2n^(n-3) spanning trees of K_n that contain a chosen edge e
I don’t want a solution just a hint in the right direction
Have you already proved the formula for the number of spanning trees without this restriction?
If so: what is the probability that a randomly picked spanning tree contains your chosen edge?
Hey does the way I've formatted this equation make sense? Reffering to the boxed one.
The summary currently adds many more terms than is needed but I'll try and get that lower later
It looks like what you really want is just to sum n-H(i) for those i where it is positive?
That would be clearer as $$\sum_{i=1}^{n} \max(0,n-H(i))$$ than with the absolute-value gymnastics.
Troposphere
Ah thanks, Yes I only want the sum when it is positive
Hmm haven’t thought of going the probability way I’ll think about it thanks
Oh I think I might see what to do thank you
Hi. I don't really know if this counts as discrete math or not... It looks like a problem in discrete math, maybe... Say I have a list of numbers $\theta_1, \theta_2, ..., \theta_k \in [0, 2\pi)$ in (strictly) increasing order. I know that $0 = \sum_{j=1}^k e^{i \theta_j}$. I'd like to show that $e^{i \theta_{j+1}} = e^{i \theta_j}e^{i 2\pi/k}$ holds for all $1 \leq j \leq k$, where $\theta_{k+1}$ is to be taken as equal to $\theta_1$.
phao
For k=2,3 I know how to do it... but what I did doesn't really generalize...
(btw... idk if this is true or not -- I think it is, but I'm not sure)
It is not necessarily true for k>3.
right
We can reformulate it as: Given a convex polygon where all sides have length 1, is it necessarily regular?
Yes if there are three sides, not if there are more.
Ok.
For example we could have theta1=0, theta2=10°, theta3=180°, theta4=190°.
Yep... when you said it
that wasn't true for k > 3
I thought of combining "two working cases for k=2"
tbh... idk if it would work, hehe, I'm just assuming that is what you did.
But thank you!!
Essentially that gives the same solution as I have. I was thinking of it more geometrically, though.
Right. Thank you very much. You really helped me here hehe
jo_fish
I am sorry to hear that
I'm kind of new to sets, am i stupid or are none of these actually correct
if you do the math in A it'd be {3, 3, 3} and B would just be {2} right
sets don't have repetition so A={3} and B={2}
But yeah, something's off here
As it stands none are correct
If you were to be pedantic,A inter B is literally null
Well ok, Nothing matches even if you don't assume that
can I get help with the inductive step
I have case 1 covered
but idk how to go about case 2 when b=0
You haven't defined what b is
2a + 7b
Wdym by 'when b = 0' though - surely the point is you need to find such a and b
when b is 0, then a is at least 3, and u can replace a with (a-3) and b with (b+1) since 7 = 6+1
why did u do a-10?
just to keep the template of 2a
its possible for there to be a case where a is less than 10 though
for example the base case, where a = 3
just do n = 2(a-3) + 7(b+1)
Oh yea, thx
I'm looking for an efficient algorithm to create cayley matrices, to be more precise:
I'm looking for a randomised algorithm that can efficiently colour every vertex of a dxd grid with d colours, such that every colour appears exactly once in every row and once in every column. kind of like a sudoku generator in more dimensions, but without the small box constraints.
I'm not sure what the best way is to go about this, since I want d to be very large and at the end I should just have d matrices of that size.
an additional constraint that isn't necessary, but one that I would possibly also like to have is that the colorings of the grid are symmetric with respect to transposition, but that is of lower priority. I can just symmetrize the matrices I got from before, worst case scenario
ok what I might just end up doing is shifting the identity matrix diagonal step after step, and then at the end just multiply the all matrices by the same random permutation matrix at the end
not sure if this generates all of them, but this should be good enough
guys, in automata and formal language theory, what would be the result of this: {{a} U {b}}*
- for Kleene Closure
The set of all words on the alphabet {a,b}
I mean, by definition, the set of all words on {a,b} is {a,b}*
would it include the empty word
Thanks man
yo sweet this server does discrete too?
I might come here often this school year too
Is my understanding right?
If I can picture properly in my mind, the Cardinality of sets A, B and C (only respectively) remain the same no matter if we change the Cardinality of the intersections
The expression looks right.
I don't have enough context to comment about "the cardinality of A, B and C remains the same".
Cardinality of elements in A only, Cardinality of elements in B only, Cardinality of elements in C only remain the same (and are equal) even if we change the Cardinality of the intersections
It's certainly possible to change the sets such that the cardinalities of each set stays constant while the cardinality of the intersections change.
I don't think you have a described any particular assumptions about the (so far unspecified) change that would force that to happen.
im trying to define an equation and am using set builder notation for part of the definition, i want to say x is an element of a sequence, but im not sure if I'm allowed to use ∈
I said p ∈ w, and defined earlier that w is a sequence, is that sufficient?
sounds a little dodgy
can you show the full definition you're trying to put into notation?
in plain english, if possible?
C sub i, where i is the index of a word, is equal to the tuple (p, x) where p is an element of the sequence w (where w is all words in the excerpt), and x is an element of the sequence i (i contains the index’s of all words in the sequence w), and if the cosine similarity of the embeddings of w sub i and p is greater than .8
Sorry if thats hard to follow
I guess i can assign w to another variable and u that as a set in this scenario
Hmm, I would expect "plain English" to imply "without assigning letters as names for everything".
It doesn’t necessarily need order
Ok yeah ur probably right
Need to create a sequence of sets, the set at each index is a word in an excerpt, containing the same index. Next get the word embedding of said word and treat it as an anchor word, then get the word embedding of all other words in the same excerpt and compare it against that anchor, if the cosine similarity is above .8 u add it to the set
What is a "word embedding"?
Basically a list of numbers that represent some features, in this instance it’s derived from the weights of a neural network called word2vec
It encodes features into those numbers, and lets u do traditional math on words basically. The traditional example is king - man would output queen
And u can go further and compare the 2 vectors to see how similar the 2 words are
does anyone know how to prove this proposition "∀x, y ∈ R, if x is rational and y is irrational, then x + y is irrational." using the definition "∃a, b ∈ Z, a = bx and b /= 0"
I'd try for a contradiction. Assume x+y is rational, see where that takes you by applying your definitions and stuff.
(See what happens if you solve for y.)
Can anyone direct me to a resource to learn how to do problems like this
my professor explained it in such a bad way
im very confused
oh wait if ur still here, is y supposed to contradict my assumption since I assumed that its irrational and I got it to be rational?
so then my assumption is wrong?
Yeah basically. You assume x+y is rational, show y must be rational too. But you had already previously assumed y was irrational. So you found a contradiction.
So, by contradiction x+y can't be rational.
Ohh ok perfect and also I set y = sqrt(2) but for x I just subbed it in as a = bx. Is that fine
so then I made m = ny which is m = sqrt(2)n
just double checking to see if im supposed to do that in my proof
Why do you think y=sqrt(2)?
What if y is some other irrational number?
oh yeah true, idk I just assumed it to be sqrt(2)
but ig I could leave it as y and not specify it
Well the proof won't be correct if you only do it for y=sqrt(2) because the thing you are trying to prove needs to hold for any irrational y.
Suppose you have k length list of objects to choose to permutate from n objects. Not all these objects are distinct. By this I mean we can partition these n objects into m distinct "types" (1<=m<=n). How many ways are you able to permute a list k long out of n objects with there being m distinct "types" in the set of n objects
To add to this I will supply this partition set function that tells you the amount of each type are made up in this n long list.
Basically this question stems from me wanting to general an example such as
AABBBCCD and permute/ take combinations of length 2
if m =1 then all the objects in the n long list are the same, if m=n then all the objects in the n long list are unique (idk how I handle the cases in between) (plz ping me if you know how to do this in general or if you have some resources that says you need to just brute force it and there is no nice math formula)
<@&286206848099549185>
Hello. What does practical Graph Theory look like? What are the methods?
For example, in Category Theory (at my level) the time is split more or less evenly between trying to imagine and type check complicated definitions, assembling diagrams from given building blocks, and drafting equational proofs about said diagrams. All of these kinds of work are productive and apply everywhere.
In Graph Theory, many problems and definitions are formulated in terms of graph homomorphisms. But finding graph homomorphisms is a hard problem. For example, a core of a graph is a smallest retract. Am I supposed, given a graph, to find and test all of its retracts? Then, most of the work must be done on a fast computer? Or maybe I could look my graph up in a library of graphs on the Internet? This is made even worse by how all the nodes are equally boring. In Category Theory we commonly have a few vertices half of which are somehow special and recognizable. Not so in Graph Theory.
I want to study big random graphs, but I cannot figure out where to start. I have been opening some course books about Graph Theory — they are easy to understand, but the problems they offer are toy problems — curious little theorems and cute little examples.
Eventually I want to connect the study of graphs with the study of simplicial complices and topology. Are there any common methods to these areas? At first they look distant. I know there is the cycle and cut space, but I never figured out how to use these tools for any practical purpose.
Another way to look at graphs is of course as at homogeneous binary relations, but commonly seen relations are boring — for example, the most intricate equivalence is simply a bunch of cliques. So it does not seem to be immediately fruitful to integrate these two areas.
TL;DR: I want to do something with big random graphs, what are my options?
based on your example AABBBCCD and permute/ take combinations of length 2 you can write a generating function for it by combining generating functions for the letters. For example B appears 3 times so we would make $1+x+x^2+x^3 = \frac{x^4-1}{x-1}$ for that, the exponents represent the number of times you can pick that letter. So when we multiply them all together, the exponents will add, so in this case you'd have $$f(x)=\frac{x^3-1}{x-1}\frac{x^4-1}{x-1}\frac{x^3-1}{x-1}\frac{x^2-1}{x-1}$$ and the coefficient on $f(x)$ of $x^2$ is the number of combinations of strings you can make of length $2$.
Merosity
I realize that's kinda gross but it's better than nothing
ngl I don't really follow where the 1+x+x^2+x^3 came from
Im not familiar with this
and I don't really see how I extract information to solve the problem from this :v
might help to look at a smaller example or work it out explicitly by expanding $(1 + x + x^2)(1 + x + x^2 + x^3)(1 + x + x^2)(1+x)$ and look at the $x^2$ term
Merosity
at least to see how it works
I am more so confused on the polynomial stuff
we never did anything like this in my prob/stats class
ok so simpler example, let's say we have these two polynomials (ax^2+bx+c) and (Ax^2+Bx+C) and I multiply them together, what will be the coefficient on x^2?
the method is what matters not the final answer here
Merosity I am honestly confused where the polynomials are tying into this is my problem.
None of my permuations or combinations dealt with this
we're wanting to get combinatorial information out of it so we are thinking of choosing a term from one polynomial and term from the other
right, that's what I'm trying to explain lol
an x^2 term can come from choosing an x^2 term from one and x^0 term from the other, or by choosing an x^1 term from one and x^1 from the other
the exponents are telling us the amount of times a letter appears, so maybe I should go more concrete than this
I did a problem like this (a+b+c)(g+d+f) and we were asked how many terms are in the expression and it would be (3 c 1) * (3 c 1)
but again it didn't have repeats and stuff so 
like the process of expansion is analogous to computing all the combinations
for the factors being multiplied
anyways idk if anything I said was related to what you were saying
sort of
it might help to look at a separate example altogether to learn some of the basics
Okay why are we using x^2 what example are we working with that we are rewrittting in this polynomial
like the binomial coefficients
if you expand $(x+y)^3$ while keeping the order and not using exponents you end up with $xxx+xxy+xyx+yxx+yyx+yxy+xyy+yyy$ which is literally every string of length 3 with 2 letters
Merosity
the coefficient on x^ay^b will be exactly the number of ways to make a string of length a+b with a xs and b ys
nope no order
oh
hello
the order was just to explicitly show the strings it's counting that it simplifies down
at each multiplication you're making a choice of x or y in (x+y) to pick in contributing towards computing (x+y)^n
okay
i did this generalization intuitionally by thinking how to generalize these 2 theorems: (k₁!)(k₂!)...(kᵣ!) | (k₁+k₂+...+kᵣ)! and (k₁!ᵃa!)(k₂!ᵇb!)...(kᵣ!ᶜc!) | (k₁a+k₂b+...+kᵣc)!
is my generalization true? (couldn't prove it)
so you can think of the coefficient on xy^2 in (x+y)(x+y)(x+y) as having come from picking x from the first (x+y) then y from the second two (x+y) when it distributes
or the other 2 ways of doing it,
which gets 3
okay
the main two things I'm trying to get at is we're going backwards from the exponent to where it came from to determine its coefficient
and that comes from looking at all the separate products and picking one term from those
okay so you are saying something like
aab
would be like
choose 2
(a^2+b)^2
or is it
might be easier to explain this in voice chat with me drawing it out
cause that's not right at all
idk I am trying to figure out how the duplicates fit into this
so let's say we have 2 As and 1 B what would I make as my generating function?
(1+x+x^2)(1+x)
now the first one is for A, the second is for B, so I think of it like this:
I don't get where that comes from
I'm showing you right now
the exponents are the frequency of that letter, so we can pick A either 0, 1, or 2 times
we can pick B 0 or 1 times
so when we look at where the $x^2$ term comes from in the product we get $x^21 + x^1x^1 = 2x^2$ since there are only 2 ways to make a string of length 2 from these options
Merosity
the first term $x^2x^0$ means 2 of A, 0 of B, and the second term $x^1x^1$ means 1 of A and 1 or B, which are the only ways of picking two letters from this
Merosity
I'm not even sure if this is what you want because I didn't understand the original question but we got side tracked into discussing how generating functions work rather than whether this generating function could work for you
basically my question is this
the second answer there is basically how I would fix up my solution
that's an exponential generating function
I guess I am taking generating functions for granted here, I'd recommend looking at some of this, it's pretty fun https://www2.math.upenn.edu/~wilf/gfology2.pdf
okay well I think I have a resource. I'll look into this generating function thing
so are the solutions to these even easy to get
Like You get some generating function but then you still have to I assume find for where x^100
depends, sometimes there are tricks, other times no
the most straight forward way to get the x^100 coefficient is differentiate 100 times, evaluate at 0, and divide by 100!
so practically speaking you probably won't be doing this by hand
the generating function is considered the solution, and you can throw the thing in a computer to get the answer out of it
basically, as math gets harder getting a nice simple solution is just not always possible, but generating functions are a way of nicely packaging the information
although they can be used for other things as well even when you do know a closed form solution
can someone help?
why can we not intesect A = B both sides by C
and have them become equal
the answer says its false
Trying to come up with concrete counterexamples helps.
So for example pick C to be the emptyset
Pick A=C and B={1}.
What happens?
Fwiw to find a counterexample you would want to start with two sets A and B that are not equal.
but when intersected with C give equal sets.
Rather than starting from A=B.
you're asked to tell if it's possible to go the OTHER way
But how do you know when to look for a counter example?
do you just automatical;ly assume its false?
For me I just periodically check
Like, if I keep failing to prove something on this type of q
Eventually I wonder if finding a counterexample is easier so I give it a shot and see what happens.
Yeah. Idk if there's a perfect answer to that kind of question. 
honestly I just dont have this math vision like for this example it took me so long to realised to union both sides by a intercet c
like how does one even spot that
aka i get its practice
but still lol
Over time it can get a lot easier.
You get used to seeing certain tricks and stuff.
Also with those kinds of questions you can usually prove them multiple ways.
ya i know alot of times my method varies but i still get it right
It's a good exercise to try to prove this by element chasing too.
i have never even heard of that
Using the definition of set equality.
A=B if and only if A is a subset of B and B is a subset of A.
Is that sounding familiar?
A is a subset of B if and only if x being an element of A implies x is an element of B.
ohh ya aka not a proper subset
So a common way to prove A=B is to use this
To show A is a subset of B, then to show B is a subset of A. Then you can conclude A=B.
Sometimes people call that kind of thing element chasing is all.
Like for this.
Assume A-B=B-A.
Assume x is an element of A.
If x is not an element of B, then x is an element of A-B, so x is an element of B-A too. But then x would end up being an element of B, which is impossible.
So, x must also be an element of B.
So, A is a subset of B.
Similar reasoning will show B is a subset of A.
So, A=B.
aghh its too late for this shit hahah
but thanks for all the help
really did clear som,e shit up as far as methods go
thank!
…

idk how you guys can study that late
"late" is relative, because timezones.
past 12
Every time is past 12; it's just a question of which 12.
silly tropo, if x and y both equal 12, then x = 12 = y. so there's only one 12
I reject the transitivity of equality
I mean, just because carrots and cucumbers are both orange doesn't mean they're the same thing
Hi, What is formula for generalized counting and permutation?
n! / k!(n-k)!
is this formula for generalized counting?
n+r-1 chose r
and is this formula same as above? kindly explain difference
the first is n choose k or $\binom{n}{k}$
Lochverstärker
i dont know what generalized counting is
That second formula looks like you are talking about the multichoose function?
N choose K gives the number of K-subsets of an N-set
Whereas the latter, N multichoose K, gives the number of K-multisets taken from an N-set
Where a multiset is just a set that allows for repeated elements
maybe one that generalizes to sets with non distinct elements. So permutations or combinations of n total elements with subsets of k and possibly some weird partition of the elements into their own "types" (1 <= #types <=n). Meorsity was telling me this has something to do with generating functions
I wasn't saying that, I was just showing an example of generating functions with binomial coefficients
this looks like what you'd get from doing stars and bars if you've heard of that, since you can think of objects and dividers between them as the things you're arranging
well nvm
i still need to look at that sometime
generating functions are cool haha
anyone here familiar with fair division?
Per48edjes
Whoops, sorry for this y'all
you are cool 
could anyone help me out with proving this proposition? I'm trying to prove it using the well ordering property and I figured out all the steps except when solving for m - 1. By the way, the m is the variable I used as the smallest element in my set
well my prof wants us to do these questions using induction and wop, its practice for my midterm I got very soon
idk its weird why they ask to use both methods instead of just induction
Can anyone explain what is vertex packing for graphs in simple terms?
the vertex packing problem is the problem of finding, for a graph G, a subset of the vertices of G in which no two vertices are adjacent that has as many vertices in it as possible.
It's the 3rd semester and we have discrete mathematics
It has related topics of CS
Hi guys! I'm going to host a debate on set theory. The topic will be 'Is the Axiom of Choice true?' Let's start the discussion! Every time you have to say something, make sure to ping me. Thank you.
hello
what is the difference between the universal statement and the universal existential statment
this isn't clear for me
both are referring that all the elements in the set has a certain property
the first has two quantifiers
other than that no idea
i never heard those terms, i would wager its not too important
which are?
$\forall x \in \bR\exists y\in\bR\colon x+y=0$
Lochverstärker
something like that, two quantifiers
wow
the other is $\forall x\in\bR^+\colon x > 0$
and thanks
Lochverstärker
one quantifier
is this book in general a good one for a CS student?
i am intending to read and study it
in what terms
(and spends half the time on intro proofs instead of anything i would call discrete math)
its just long
i love proofs😂
you could cover the same content in maybe 30% of the pages
but you have to decide yourself if being wordy is good or bad
sure, thank you.
it's probably good for you, go for it have fun
Yo I have the same book, it Def is long. That's good to hear because I learned nothing about proofs in school. I'm only on chapter 2
Can anyone explain what is vertex packing polytope of a graph? (in simple language)
how does one answer this?
Let X be the set of nonempty subsets of R, and let R be the relation on X given by ARB if and only if A ∩ B ≠ ∅. Is R reflexive?
what does this mean
what does what mean
X be the set of nonempty subsets of R
this
what is R referring to in here? so confused
the ans is it is reflexive but i don't get the explanation nor the question
is it a fancy R? like $\bR$?
Ann
no it's just a normal one
show screenshot?
like i find the statement itself very confusing, R is probably referring to the relation itself
sounds like you have a conflict of notation
and X is a non empty subset of R
let's call the relation S
we have the relation "A and B have nonempty intersection"
the question is
is every set related to itself by this
and the answer is obvious: A intersect A = A, and the sets we are considering are nonempty, so yes
where is this piece of information found in the qn? How do we know that the we are checking the relation of each set with itself
that's the defn of a reflexive relation, against which we are to check this one.
ahh okk , but the whole lot phrasing of the qn still confuses me. Are A and B both elements of X?
yes
how can X be a subset of a relation which is based on itself
that part confuses me a lot
.
the question mistakenly uses the letter R to refer to two different things
that is how i interpret it
oh so the first R is real numbers?
ahh ok it make sense now, thanks man
is there any identity for $\sum _{j=1}^NjP\left(N{,}j\right)$ where $P(N,k)$ is the number of k-permutations of N
Tadashi
i don't understand the logic on conditional
Is there something specific about it you don't understand?
Is it the fact that conditional is only false for T -> F
not only that
I don't understand his logic
Yes, I'm asking you to try to explain more about what specifically you are having trouble with.
Like, are you getting stuck on why we define it the way we do in general?
like how it works
It's just a definition.
We define "A implies B" in terms of the truth table of the conditional.
This kind of conditional is called the material conditional.
There are other types of conditional.
like the inverse ecc..?
But the material conditional is the one people generally use in math.
No
like what?
Well the way I always thought of this was that the conditional was a contract.
“If my it rains, my shirt gets wet.”
The only time this statement is false if it rains and my shirt doesnt get wet.
If it doesn’t rain, then we completely ignore the conditional because im only making claims about scenarios where it rains.
Stuff like inverse, contrapositive etc are just names for certain formulas involving the material conditional iirc
But there are whole other logical formalisms meant to capture other notions of conditional statements
oh , i was confues about that
Like strict conditionals
Don’t confuse him more lol
Junk like that
so in the second case is it true?
I don't think it will
If you mean the case where it rains and my shirt doesnt get wet then that’s false
I look at the material conditional as kind of a "good enough" approximation of what "A implies B" type statements mean for the purposes of doing stuff like math.
If it rains my shirt must get wet. That’s the contract
The other one
For remembering the truth table definition of the material conditional all you really need to know is T->F is the only case where the conditional is false.
If it’s a sunny day and my shirt is dry or if it’s a sunny day and my shirt is wet that doesn’t break the contract.
It’s a scenario “outside” of the contract. Remember I’m only making a claim about scenarios where it rains.
Briscola do you know the definition of the conditional we're talking about?
^
Yes
Mmm idk if that truth table came out legible lol
Yes it is
I mean that
I understood
Thank to @unique abyss
Okay, yeah. So this table is basically the jist of the defn of the conditional.
So you’re not alone by any means
I usually think of the intuition for it in terms of real life examples.
Like, if I promise you "if you get an A on your test, then I'll buy you pizza."
If you don't get an A and I don't buy you pizza I'm not lying.
If you don't get an A and I still buy you pizza I'm not lying.
If you get an A and I do buy you pizza I'm not lying.
If you get an A and I don't buy you pizza we have a problem though.
I told you I would buy you pizza in that case, but I didn't.
So it seems like I've lied to you in that case.
That's the only false case in the definition of the conditional.
It's the T->F case.
@rancid pivot @sour onyx i dont wanna clog that channel
if you know that the four is on the right in particular, you know that the third visible face has to be 6
So say you can see 5 and 6. The third side is 3 or 4.
yeah that i know
No, the third could be 1
imma just hardcode it probably lol
i meant like orientation
cuz yeah if you just know which ones are visible there are two options
but if you know on which face in particular they are, you can fully determine the third
idk how to do that tho
Unless you know the exact orientation of the die, there's no way to tell which of the 2 numbers it will be
Hm say you know 6 and 4 you know opposite six is one and opposite 4 is 3 if we conclude that even and odd sums of adjacent sides must touch (based on this photo) hmmm how do I describe this better
oh shit i see what you mean
For a balanced d6, you can take any two opposite faces and switch them and the die will remain perfectly balanced.
say you have the top and the left face
if the sum of those two is odd then the third face must be even
and vice versa
Okay so like looking at the 4, to the left is six the sum of those two is 10 which even, on the top we have 5 which summed with 4 is nine which is odd, then the right is 1 and you get a sum of 5 which is odd and on the bottom I believe there is 2 which summed is 6 which is even
kenshin i love you
Right but there are still two solutions to that
Think of the die in terms on adjacency and opposition. Choose on opposite pair. Each member of that pair is adjacent to the same 4 sides. If you switch the places of these opposite sides, they will still be adjacent to the same 4 sides.
You need a stricter property than balance if you want there to be one answer here
right but im only worried about non-opposite pairs
OK let me phrase this another way. There are two kinds of balanced d6. You can show this by unfolding the faces. One type is as follows:
6
2 3 4 5
1
1
2 3 4 5
6
If you can see 3 faces like in your picture, where the left face is 4 and the right face is 5, then the top face can be 1 or 6. However, if we have the first type of d6 then it must be 6 and if we have the second type then it must be 1
Unless you choose which type of balanced d6 you're working with, you cannot definitively say which side is there.
ohhhh you mean like a variety in the actual way the dice are made
yeah i dont care that much about that as long as the results i get consistently agree with only one
im just going by the die i have lol
which seems to be the second type
i see now what you meant @rancid pivot
took a break and came back and it doesnt work lmao
Lol glad I could help!
the problem i ran into is that like even with just the die i have
going in order of top,left,righ - (6,5,3) is one ordering but (5,6,4) is another and the sum of 5 and 6 is obviously odd but it doesnt determine if the last face is 3 or 4
i couldve hardcoded this by now 
welp this is no longer a bug, i hereby declare it a feature
Welcome to software engineering lol
do you know if it's possible then :(
cuz like the kind of die im using is fixed
just realized king dice is actually a 1 in his face lmao
Can you help me find the partitions ?
I know the ans but I cant solve it by theoretically
Tag me when you uh solve this
,rcw
I also think part of the issue is that it’s not exactly clear what the “third face” is
The third face has to be adjacent to the two given faces
That leaves two faces as possibilities
Based on the way you’ve described the problem, among these two faces, it’s the one visible to the viewer
I think for hard-coding
Let’s see
So every face has four adjacent faces
And there are 6 faces
So 24 total pairs I think
And if 5 is on top and 4 is on the right that’s different from if 4 is on the top and 5 is on the right, if you include the fact that the third phase must be “visible to the viewer”
So the ordering of the pairs matters
I need help building an understanding of strong induction
The difference between strong and weak induction is relatively minor
So if you understand induction in general, you’re 99% there
I think I understand normal induction and I understand the idea behind strong
But I don’t know what it is
Like I understand that weak induction works when you don’t need to assume it works for all k < that value right?
Yeah
And strong induction is when you assume that it holds for values between the base case and the k-th step inclusive
How do you prove that it works up to the k step?
You don’t AFAIK
It’s assumed that it works up to k step in strong induction
so how is that different then normal
you just assume?
dont you need to do something to get to the assumption?
do you just prove 2 base cases
instead of the first one?
The difference lies in the fact that you don’t need to assume that the statement holds for all values from base case to k
Weak means that a statement for n implies n + 1
In strong induction, you are assuming something extra
That the statement is not just true for n
But all values between base case and n
you dont need multiple base cases
the difference is with normal assumption you only assume the previous domino has fallen and show the next will fall and with strong induction you assume all previous dominos have fallen
there is no issue here because strong induction is implied by normal induction
you prove your base case say P(1)
and if you have proved P(n) -> P(n+1) you can use this repeatedly
then you get P(1) -> P(2) and P(2) -> P(3) and so on
actually this explanation isnt best, hmm
(the dominos one is fine)
I understand the dominos part so how can we make that assumption?
ok
say you prove something by strong induction
so you prove P(1) and you prove P(1) and P(2) and ... and P(n) -> P(n+1) for all n > 1
so what you did is prove P(1) and P(1) -> P(2) and P(1) and P(2) -> P(3) and P(1) and P(2) and P(3) -> P(4) and so on ...
but each step can also be proven using weak induction just more often
so say you only know P(n) -> P(n+1) for all n instead, call this (*)
and you want to prove P(1) and P(2) and ... and P(n) -> P(n+1) for all n
then what you can do is apply (*) to show P(1) -> P(2), so you now know P(2) is true
Ok I see the difference
But does weak always work?
Or why would they teach strong
so you can continue doing this and keep applying (*) to get the strong statement
strong induction is better
you get more assumptions during your inductive step, which is well, stronger
often you have some kind of object that consists of n "pieces" and you want to show by induction that it has some property
so a valid strategy is to break up this object into two smaller objects consisting of n-k and k pieces and use the induction hypothesis on either piece
and then piece together that information somehow
now if you only have weak induction this doesnt quite work, you can only apply the induction hypothesis to smaller objects consisting of n-1 pieces
if you have strong induction you can apply it to any object with less pieces
now you can turn every strong induction proof into a weak induction proof but that hassle isnt worth it
the better question should be "why do they teach weak induction"
And note that for proofs where weak induction is used, you can still use strong induction. Strong induction still assumes n-1 like weak does. Just with some more assumptions in between
I get a lot of what you’re saying here but I still don’t understand how to make the assumption that it works how would you structure the proof differently to prove up to k
the structure is not different, you just assume more during the inductive step
i agree my explanation isnt great, i kinda lost my train of thought lol
You don’t need to do anything differently to prove it?
maybe the top answer here helps: https://math.stackexchange.com/questions/108297/strong-induction-proofs-done-with-weak-induction
To prove assumption
no
strong induction is kinda just weak induction applied repeatedly
(though a finite number of times)
the top answer here shows this kinda
if Q(n) = "P(k) is true for all k=1, 2, ..., n", then weak induction is show P(n) -> P(n+1) and strong induction is Q(n) -> P(n+1)
but if you prove P(n) -> P(n+1), then you can use induction to prove Q(n)
so everything works out in the end
Hmmmm
Here I have a question from class
Why does it ask to prove a2 a3 and a4
Is that relevant for induction or just a random thing to show you understand how the sequence works
,rotate
Looks like fibonacci to me
It was an old midterm question
It's just to show you how sequences work
it looks like (a) just exists so you convince yourself that the statement you have to prove in (b) is true
or yeah, that
or just easy points
@plucky island If you're willing to read alot, you may be interested in this answer https://math.stackexchange.com/questions/1184541/what-exactly-is-the-difference-between-weak-and-strong-induction
top answer
it has a rigorous proof for strong -> weak and weak -> strong
Is the set X = {3a + 5b | a, b \in \mathbb{Z}} equal to \mathbb{Z}?
hint: can you make 3a+5b=1?
Yes, hence taking some multiple of that will yield any element in Z, thanks 👍
nice
ok now generally what conditions on integers u,v are required so that the set X = {ua + vb | a, b \in \mathbb{Z}} equal to \mathbb{Z}?
Hmmmm, u is coprime with v?
yep that ends up being the answer
Okay, good to know
Thanks (if it wasn't obvious, this is part of a bigger question, just a spurious thought 😄)
Amorous aka Lucifer
oh, that kind of set theory? #foundations may be better suited for it
is there a reason why the word "that" is in italic?
thanks
italics are the plaintext equivalent of verbal emphasis
when you pinged me in #precalculus asking where to post set theory questions, i thought you might have been referring to more elementary stuff like set counting with inclusion-exclusion
and the like
ohk
got it
I'm really lost.. I really don't know where to start. The question is
A rectangular chocolate bar is made up of n squares. The whole bar or any part of it
of the bar can be separated along a vertical line or a horizontal line which separates the ´
squares. If the bar can only be separated one part at a time, determine the division number
that must be done in order to separate the n squares. Use the strong induction to prove
your answer.
Why do i need the strong induction ?
maybe because weak isn't working
what is your approach?
Is there a channel for Curve Tracing, asymptotes, radius of curvature
Amorous aka Lucifer
and yes you don't need strong induction...it's just because you have been told to do the problem in a certain way by your instructor
You are using strong induction above though? Im a bit confused
Hmm, is there an identity which states that if a \subset b and b \subset a, then a=b?
that's the definition of a=b
Oh, cool
or well an equivalent definition anyway. who knows how your course defined it
i am using but its not necessary
Suppose, I have an undirected planar graph of n vertices, what is the maximum number of edges possible?
every edge corresponds to an unordered pair of vertices
how many of those are there?
n choose 2 ?
is that a question or an answer
i doubt whether it's the answer
why do you doubt it?
yes it is
graph theory is just weird to me
however, i can understand and apply it on the questions
still, why do/did you doubt it?
Hey I gotta induction question can you prove something assuming k-1 and then prove k
It’s logically equivalent yeah?
Yeah, this is typically what you see in induction
and then k implies k+1
and k+1 implies k+2
thus creating a chain of implications
that's why induction works as a proof technique
Gotcha so you can either assume k and prove k+1 or k-1 to prove k whatever makes the question easier ?
And if you have strong induction with your assumption being 1<i<k can you just prove k from i or do you need to still do k+1 cause logically isn’t it the same as k-1?
For strong induction, your assumption is that the statement holds from base-case to k inclusive and then proving k -> k+1
the only difference between weak induction and strong induction is the assumption
I see
Does that assumption ever let you do i as a k-1 term though
or no
have i and then prove k?
can you re-phrase that
If you're saying, assuming base-case to k-1 inclusive and then proving k
sure you can do that
That could be considered strong induction
So let’s say we have a function that we can satisfy from 1<i<k could you have the function of i be equivalent to a k-1 term then prove k
Does that make more sense
Maybe I can look for a question
Here I found something where I think that’s what they did
It's strong induction yeah
Idk why it's assuming the statement for 0 ≤ i ≤ k
and then proving the statement for k
when it's assuming that in the inductive hypothesis
Maybe there's something I'm missing so I'll defer your question to someone else
Okie well you’ve been good help I’m kinda confused bout it also
could just be a typo and that they mean to write 0 ≤ i ≤ k-1
Yeah I’m not sure
can someone explain how to find equivalence classes for this problem
our professor never went over this and he gave it to us as a hw problem
btw I already showed it's a equivalence relation
The problem asks you to split up A into a disjoint union of equivalence classes. Recall that for any a∈A, the equivalence class [a] is the set of all b∈A such that aRb. {-3,-1,1,3} and{-2,0,2} are the equivalence classes. (In an equivalence class each element is related to every other element.)
how tf did you get that tho
well you could for example just start with one element, e.g. -3 and go through all other elements a in A and check whether -3Ra
then when you are done with that take the next element which isn't already accounted for and repeat
in this case you can speed it up a little by calculating a^2 mod 4 for all a in A and noticing that aRb iff a^2=b^2 mod 4
Hey guys, any chance someone knows of a proof or intuition video (well or text) for proving the connections between Zorn's lemma ~ Axiom of choice ~ Well order principle? but with all 6 proofs? 🙂
Hrbacek/Jech 3rd edition ch 8 covers it if you can't find a video.
Hey guys. Is anyone here familiar with Sudoku Algorithms? I have a puzzle I'm working on and could use some help. It's in a Sudoku like format and I suck at discrete math. I think some First Order Logic is needed.
Every time I try making expressions they end up stupidly long. Just the first cell ended up taking half my whiteboard.
Hello, I am trying to do some math homework and I am stuck on one step. I understand all the way to the associative law, and I know the correct next step is that distributive law, but I don't know why. I can't work out in my head how that was reduced that way using distributive law
this is my distributive law that I am using, I can't see how either one of those are working here
that step actually uses two instances of distributive law
ya I figured it was using one on the right and the left
I just cant figure out what they are plugging in for p,q, and r both times
the thing being plugged in for p is the term appearing twice
slash the thing occuring outside the (nested) parentheses in the line before it
hmm, are they using the first or second law?
you tell me
one wedge/vee occurs inside the parentheses and one occurs outside them
^and the other
(wedge = /, vee = V)
discord deletes symbols I guess, wedge = upside down V
so wedge is the AND symbol?
AVA
still figuring out discord
I don't know, I feel like my brain is melting
I can't find a repeating P in there, it's probably something simple I am missing
the first PˬQ I can't tell how it's turning into what they have it turning into
complex expressions can be brain melting indeed
if you've got a notebook handy, I recommend just writing down the expressions inside the leftmost [] brackets
one on top of each other
and it'll make it easier
and then do the same with the rightmost [] brackets
what's messing with me is that hte left most brackets only have two variables
and the distributive law has 3
or are they treating ¬q as an r?
They are
You can always substitute a more complicated expression in for a variable
i.e. we have the commutative law x+y = y+x, and so you can say (2z+xy) + (3y-z) = (3y-z) + (2z+xy)
and you can re-use the same variable too, i.e. "x+x = x+x"
or maybe a better example is going from "(x+y)+z = x+(y+z)" to "(x+x)+y = x+(x+y)"
it's this mcgaw hill connect thing, where I place all of those in proper order and go to the next one
they have some unuse dones
I can work out the right answer by elimintation and logic but I wanted to really understand each one fully
Can I find it online?
You should definitely not search for it online on piracy websites. Definitely don't illegally pirate it and read it for free.
Ok thank you, I'll go look for it on amazon 🙂
you can do a standard "let x be in [LHS / RHS]... then x is in [RHS / LHS]"
any luck with either direction?
I got a different answer to the solution but it seems valid.
Sum of number of correct answers is 720 or more so atleast 1 person who solved 4 or more questions. Say this person solved Q1,2,3,4 (Without loss of generality) then the sum of the number of people who solved 5 and 6 is 240 so at least 1 person in their intersection so we are done.
Hey people, can some check my work. It is a question on permutations
The question is on permutations:
How many 4 letter words can you make using the word B A N A N A S? repititions/duplicates not allowed.
that doesn't look right to me
this assumes all selections of 4 letters will feature the 3 A's (hence division by 3!) and both N's (hence division by 2!) which just is not true
(also you are inconsistent about stroking your 7's; all of them should have a stroke, not just the first one on the page
Isnt AAAN and NNAA part of the 840 ways?
Ill take care this time
yeah but they arent the only ones
so you cant just divide by 12 like you did
Is it correct till 840?
well sure but you can't really proceed past that point anyway
also just saying. this is not a seven. this is a one.
I have 7 letters (n=7)
I have to arrange them in 4 ways (r=4)
Damn that is bad, ill make sure that wont happen the next time
i am critiquing your handwriting here.
you should always, always, ALWAYS put a stroke through your sevens.
every time you fail to put a stroke through a handwritten 7, a puppy dies
anyway
one would think that we must go through some casework here
perhaps baesd on whether or not B is included and whether or not S is included
or something
i'm not yet sure what the best case breakdown would be
but i see no way to avoid one
I dont get it
Ok so you say till 840 it is correct yea?
So i have 840 4 letter words i can make with BANANAS
no
But the 840 has repitions in them
you have 840 four-letter words you can make with $BA_1N_1A_2N_2A_3S$
Ann
and no you cannot account for these repetitions with a simple division
So this isnt a newbie permutations combinations question
this is not a newbie question, no.
Where i divide by the duplicates and get the answer
Cuz i just started off with permutations and combinations and i found this question browsing the internet
It dint fit anything i was teaching myself till now
Anyways that also means my intuition isnt correct.
Thanks anyways @stray reef
I am trying to figure out why those two are showing green as right here
The "Some" symbol at the beginning has me thinking hte top two are right
like can't there be a rabbit that does hop and one that does not hop and that statement still works?
I am wondering if this thing is just wrong or maybe I don't get something
1st is right
2nd isnt; the statement isnt talking about an animal thats necessarily a rabbit
it is though, the 2nd one says the animal has to be a rabbit and it at least once hops
i mean the problem statement
seems to fit the SOME in the beginning, again I might be missing something.. But you agree the answers they are giving are wrong right?
Well R(x) makes it a rabbit no?
even if the domain is all animals, the function limits those to rabbits
but thats not all the problem statement says
I am getting SOME function of domain is rabbits therefore hops
it says there exists x such that:
if R(x) then H(x)
if the animal is a rabbit then it hops
but the backwards E means at least once, not all
or you mean the 2nd statment would need an AND symbool
not an if/then
ok that makes sense
the question feels like its looking for 2 answers but only the first one is correct right?
same with this one, the two red ones are the ones I had as right
yes
reds are right
who wrote these?
ok cool
it's some program, that mgraw hill connect thing
I emailed my professor about it
they make you pay for this crap...
hopefully u get points back
ya I am sure I will, my professor is pretty cool
