#discrete-math
1 messages · Page 102 of 1
Take the negation and apply Demorgan's for all of it
,rotate
nvm i got it lol
Congratulations
Thanks to you guys 😄
Hey, on my first problem set I have a few exercises like this.
Where I need to prove equality. I have an idea to prove it by induction but i don't really know how to use Big O notation to do this .-.
by induction?
It is wrong idea ?
D:
Don't get me wrong, i don't want whole proof, just some kind of starting point
start with the definition of the exponential function
i assume this problem set is about introducing you to big o notation?
yes, exactly
when saying x mod y
if x is negative
what happens ?
will the remainder also be negative?
or was there a rule that said its positive ?
remainders are generally given to be a number between 0 and y-1 inclusive
you can always add or subtract multiples of y to your number to force it inbetween that range
-1 mod 5 is the same as 4 mod 5
as an example
@obtuse lance I'm supposed to find c in c ≡ a^3 - b^3 (mod 13)
her a is 4 and b is 9
-665
-665 mod 13 = -2
but I also know that 0 <= c <= 12
I just answered what to do, reread what I wrote
not sure what you mean by multiples of y
a=b (mod n) iff a-b=kn
so -2=n-2 (mod n)
Would this be a valid proof
except for notation, yes
no
you seem to somewhat use ≡ and = interchangeably
there is one line where you state a congruence but not modulo to what
Would writing
n^2 ≡ 0 + 0 + 1 ≡ 1 (mod 4)
and n^2 ≡ 0 (mod 4)
be correct?
then
yes
can someone explain to me what 4) is asking
<@&286206848099549185>
the wording is pretty weird
hmm
👀
@alpine goblet please wait 15 minutes to ping
okay it does reference to 3
Describe m as a function of epsilon and k such that...
$m(\epsilon, k)$
Element118:
so that a message with length m symbols having k symbols 1 is generated with probability < epsilon
get it so far?
yeahh
so, what have you tried?
okay
can you express the probability as a summation?
would someone be able to help me find a formula for this sequence
the 0 starting in the beginning is throwing me off
er ... separate the numerator and denominator.
Put the two patterns together with a fraction bar 🙂
well so, i know the top part is like adding odd numbers 1,3,5, but how do you get it to start with zero if n has to start at 1?
so ... what's the next odd number below 1?
ohh ok -1
do you need a recursive formula or a explicit formula?
explicit
so ... fun fact ... what kind of numbers are 0, 1, 4, 9?
Yes, you add odds to get them, but like ... how do you get from your counting numbers 0, 1 , 2, 3, 4 to 0, 1, 4, 9, 16?
yeah my brains kinda turned off at this point this was a stupid question sorry lol
perfect squares
yeah 🙂
appreciate it
Do you have to start with a1 or can there be an a0 ... oh okay 🙂
i do have to start with a1, but i think i know what to do
sure sure ... as is said by Bill Nye 🙂 ... carry on.
hey guys i just started learning discrete maths in uni and I'm not sure if proof i have written is correct
Show that log7(n) is either an integer or irrational, where n is a positive integer.
my solution:
assume log7(n) is rational > log7(n)=(a/b) > 7^(a/b)=n > 7^a=n^b
therefore if log7(n) is rational. then n must be factor of 7
is this correct solution?
,help
A brief description and guide on how to use me was sent to your DMs! Please use ,list to see a list of all my commands, and ,help cmd to get detailed help on a command!
yes thats what i wanted to write :DDD
and how would i show that a/b is an integer?
I'd do cases. When is it an integer and for all else it must be irrational.
oops nvm sorry element
or consider a factor of n that is not a multiple of 7
okay thanks for your help
i will probably frequent this text chat in this semester :DDD
Hi all, I am struggling to think of an example where set operations (i.e. union, intersection, difference etc) are used in real life or business organisations.
Have you ever heard of sql?
Sql is basically modeled off of set theory operations
@faint narwhal how are set operations used in SQL?
Well sql uses predicates, which are a common thing used in set theory
Depending on what exactly you use, you can do things like cartesian product, intersection
All of which are just set theory operations
I am not exactly clear but are operations used when writing SQL queries?
Yes
@faint narwhal Thanks, I will read a bit on SQL then
Sure
If you plan to continue to study math
You're going to get to things that have no real use or connection to the real world
Exams have just ended 😄
Recommend me something related to graphs
I want to learn them for Algorithms
I've not started with them yet^
The concepts for trees and the traveling salesman problem are widely understandable.
Sorry. I meant those three as suggestions.
hmm, graphs hmm
graph automorphisms
np complete problems on graph like 3-colouring
planarity checking
you can find problems, prove they have a polynomial time algorithm, but have no idea what the algorithm is
all because of this
as well as maximum independent set, matching, bipartite matching, graph traversal
topological sorting
@cerulean ore
Hello
someone here?
So someone told me that you can find out wether psi is tuatology by using a method called backward method. and they said that in this method, you try to falsify tautology and if there is a contradiction, then it means it is a tautology. it is similiar to truth tree
I dont know how backward method works like?
there is a channel bout logic
Discrete Mathematics and Its Applications 8th Edition by Kenneth Rosen @patent rover
Also this is a dumb question but is (p OR q) AND p equal to (p OR p) AND (q OR p)?
Help
@coarse gyro yes, p or p == p
Pls
what have you tried
I tried asking “will the dude on the left stop me if I’m going to certain death” but that is implying that the person cares lol
Idk
have you solved similar questions before
or seen them solved
you should spend more time on it, not ask for help
because the answer is simple and will completely spoil the fun away from you
@weary tiger
So the person you ask will always "care". One of the two will always tell you the truth to your question, the other will always lie.
It's a magical Island
hmm
I can find the correct answer pretty quickly with a computer
I write a computer program which is compiled
argh, too many numbers for 64 bit integers
argh the asymmetric range is causing problems for me
I'm going to deal with the lower bound separately then
argh precision is annoying me
ah, integer division
It would be very difficult to do this "mathematically" cuz integers
Oh wait, are these allowed to be reals?
big = pow(2, 61)-1
small = -pow(2, 61)
total = pow(2, 124) # all the possibilities
ways = 0
x=1 # try it out for different x
while x<=big: # for all x
maxAmount = big//x # find largest positive y, and smallest negative y
maxX = min(big//maxAmount, big) # find largest x that has the same range
ways += (maxAmount*2+1) * (maxX-x+1) # add in these solutions
x = maxX+1 # update x
ways *= 2 # (x, y) works, so (-x, -y) works
ways += 1 # unless x=-2^61, y=0, that's not included
ways += 124 # ways to make -2^61
ways += pow(2, 62) # ways where x=0
print(ways/total)
Computer is by far the best you can do here. There won't be a closed expression
I'm checking my code, the idea is there
I have 4.336808689942018e-19 of it being in the range
9223372036854775931 ways
can someone independently check?
argh
wait
I typed wrong in my code
no wonder the number looked familiar
it's definitely taking a while
while my code is churning out the answer, anyone else can double check my code?
(python)
I can extend this problem to the reals. Plus if we only consider positive numbers, we're interested in
The numbers where x ≤ z/y, for z = 2^61
∫ z/y dy for y between 1 and 2^61
= 2^61 ln(2^61)
And the area of the square, that is (2^61)² - 2×2^61
So the answer is approximately
2^61 ln(2^61) / (2^61)² - 2×2^61
≈ ln(2^61) / 2^61
Problem is, I don't know how approximate that answer is
Clearly 0.999... for like 17 decimal places
You sure? At best, that's good enough to show a computer sim is approximate
It could be off
You might be able to use some error theorems to find how far off it is
What'd you get? I'm interested
calculation would take a bit
it's python
It's looping 10^6 times a second
So, it's going to take about 2000 seconds
or an hour
great news
it's done
I count 202620924671441687225 ways (with nonegative x)
so, the answer is 9.527190092386719e-18 of the result not overflowing
@sour arrow
@gleaming dew
I got 1.833E-17
With my approximation. Jury is out whatever the comparison might mean
Wow, grats on the program literally counting the number of ways to make x, y. That's impressive
let's see
mine is off by a factor of about 2, for some reason
let me check
ah no wonder
because I didn't consider negative x
ah
don't worry, quick calculation
okay
updated
400630163324455986423 ways
1.8837539750276337e-17
@sour arrow you got 1.833 or 1.883?
The following error occured while calculating:
Error: (intermediate value)(intermediate value)(intermediate value) is not a function
That was my approximation by extending to the reals. It won't be exact
damn
nice, only 2.7% off
that's wild
Agreed, I'm suprised we ended up that close together
nah, it was more or less to be expected
now let me put in the correction factor for the reals
x=0, 1, all y are possible
x=+-2, 1/2 of y are possible...
x=+-3, 1/3 of y are possible...
So, the answer is just 1+2*sum(1/n until 2^61)/2^62
but, then the terms after 2^30.5 are going to get pretty inaccurate
so, there we flip x and y
,w 2^622(1+2*(ln(2^30.5)+euler gamma))-(2^30.5)^2*4
argh not getting the answer yet, hold on
so close, off by a factor of 2 but I don't know why
wait
yeah that looks right
whoops
let me get more digits
4.0063016332445566735637254554719563225420935286099713... × 10^20
Actual:
400630163324455986423
Estimate:
400630163324455667356
good enough?
now let's calculate error bound
assuming our calculations of sum(1/n) were precise...
ah, the fractional error for them is less than 2^-30.5
okay, then the absolute error of our probability would be on the order of 2^-92.5
@sour arrow
what do you think?
I made a calculation that has an error of only 330k
I don't understand it! :D
okay, the main idea is because if we use the entire 1+1/2+1/3...+1/2^61
then the terms at the end will have lots of error
that's were all the error came from
so I decided to use half of it and move it around
okay let me demonstrate with a picture
So, we get the shaded region
and then we place that region on the board 8 times
that's the main idea
Whatever you guys are doing looks interesting/fascinating but, alien to me.
I have no idea why it gave me so many correct digits
what... are you even doing
I can still assure you that you're a genius!
@stray reef want me to try to explain?
Given two integers, x and y, each randomly chosen from the range of -2^61 to 2^61-1. What is the probability that these two numbers, when multiplied, will not produce a third number, z, (z = x * y) where z is in the range -2^61 to 2^61 - 1 ?
Here's the original problem
oh oof
yeah, I used code to find what I think should be the answer
let's skip that and estimate the answer
The answer is near 1, but we are finding 1-answer
Now look at this diagram.
I'm going to try to estimate the number of numbers in the shaded region
the square in the middle is from -2^30.5 to +2^30.5
the region I'll count from x=1 to x=2^30.5
all okay?
@stray reef
okay, maybe next time you can bring it up
@reef thistle
Sorry for the spacing (university sends like this only)
I want to learn them (Graph theory, trees) properly as they're involved in Data Structures and Algorithms as well.
Which book do you recommend?
H OLy fucking shit the spacing is an eye-sore
@stray reef lol, could you please recommend?
recommend what
there's no way i'm gonna recommend you anything based on this sickening list of poorly typed buzzwords
oof
That's what I get from my university. Do you want me to fix the spacing?
That was rude. (Tell me you don't care)
Here's a book https://cpbook.net/
Thank you!
@reef thistle no,sir,only this https://www.pdfdrive.com/competitive-programming-3-e32649251.html
@opal ember You're also into it?
oh, thank you!
when you have this in english language and you want to represent it into a truth table , we use xor and not the normal or right? because the sentence is missing a "or both"
i see thanks^^
and if i have "if albert comes , so do cornelia and mr meyer" , if mr meyer comes , does it imply that albert came? from my understanding , it should be the case
no it doesn't
if then statements generally do not work the other way and should not be assumed to do so
okay thanks !
example:
If it rains, then the road is wet.
but if the road is wet, then it doesn’t have to be raining. It could have rained just before, or someone could have poured water there, or there was a flood…
@glossy relic
true ^ thanks
would the time complexity for T(n) = T(n-1)*c^n where c > 1 be O(1) time...?
How much time do you guys spend on a problem before checking the answer
depends on the problem
the longer you spend not knowing, the more likely you're going to remember it in 5 years from now and appreciate the solution
But, if you're preparing for a competition, then?
hows that @stray reef
c^n * c^(n-1) * c^(n-2) * ... * c^2 * c * T(0)
wouldnt it be (c^n) in that case?
sorry i suck at substitution and stuff
pretty ok with master theorem
@stray reef
@reef thistle
What say?
Till know what I think is that we have to find the letters such that one or group of them divides the numbers in this pattern
I have a pretty simple homework assignment. It is just 9 questions but worth like 20% of my grade, so any small mistake could mess me up. If anyone has an extra minute could you please look it over, and point me in the right direction if I missed something?
https://docs.google.com/document/d/1GEtf3HR3TUL9GJRUIMjyldZtDsS1Ztp46gG2lDdjc84/
Fixed the link
already your first answer has a typo, so maybe you should re-check yourself
other than that it seems fine, though
@clever nacelle
It should say is in the set of B alone, correct? I just caught that now
yeah
help
trying to prove distributivity for XOR
(r and p) XOR (r and q)
r and (p or q) and not (r and (p and q))
[r and (p or q)] and [not r or not(p and q)]
where do i go from here
btw sorry if the notation is kinda weird, it's just what I've used
I’ve seen worse I guess, the + is weird tho I’d just stick to the usual symbol for or (∨) or use sth like a pipe | because that’s used in programming langs
this is the notation usually used in ece
as far as I know
it's pretty confusing notation
Can you please help me understand what a valid subset of d would be?
(a) Is {5} є {1, 3, 5}? False. {5} is a set, not an element .
(b) Is {5} a subset of {1, 3, 5}? True, {5} is a set, found within the set of {1, 3, 5}
(c) Is {5} є { {1}, {3}, {5} }? True, {5} is an element of the set of sets.
(d) Is {5} a subset of { {1}, {3}, {5} }? False, the elements in the set are {1}, {3}, {5}.
I am sorry for the really easy question. I am just wondering. Since {1}, {3}, {5} are elements, would {{1}} be a subset?
{{5}}
so to make a valid subset, first just start off with {}
then pick any number of elements (or even no elements!) and put them in
say I pick {5}
and I pick {3} too
then I put it in
{{3}, {5}}
that's another valid subset
Alright, thank you! makes sense.
please don't solve it, but I don't understand: what is " Show that among any n + 1 positive integers not exceeding 2n there must be an integer that divides one of the other integers." asking
what's the other integer it's talking about?
sry if it's more of an english question than math :(
no matter what n+1 integers from {1, 2, ..., 2n} you pick, there's going to be a pair of them where one divides the other
that is what you are asked to prove
@gusty thorn
hm
if I chose 5
{6, 7, 8, 9, 10}
there's no pair that divides into the other, no?
2 3 4 5 5 6
sure yes 2 divides 6
the theorem says this is always going to happen. that you're always going to find a pair like this
OH
I see thank you! 🙏
n = 4
n+1 = 5
{1, 2, 3, 4, 5, 6, 7, 8}
oh wow this is pretty cool
-end-
for a binary relation, given a set A that contains {1,2}, is the relation that contains the ordered pairs (1,1) and (2,2) also symmetric? i know that its reflexive but like how can i make it so that its reflexive and symmetric but having the minimum number of pairs for an n-size set?
like say that i have a set A with n elements, if i were to have a relation that is reflexive and symmetric what would be the range of the number of ordered pairs one can get given n elements?
What do you need for a relation to be symmetric
for a relation to be symmetric both A is related to B and B is related to A, with A and B being elements
So does that relation satisfy that property
Why doesn't it satisfy the symmetric property?
if it were to satisfy the property wouldnt there be ordered pairs (1,2) and (2,1)? based on the pic i dont get why picture A would be symmetric when theres only one arrow pointing to itself
oh so the ordered pairs (1,1) and/or (2,2) can be considered a symmetric relation by itself?
im confused 😭😭
read the property again
epelta:
I need to find remainder
use Euler's theorem
what?
it's basically the souped up version of Fermat's little theorem for when you want the remainder of division without a prime, look it up
alternatively, just look at the remainder of 7, 7^2, 7^3... and 27, 27^2, 27^3... and so on
look at the pattern
@weary tiger
okay, so you solved the problem?
you only have like a small number of cases
@gusty pasture
just do casework and you are done
Do they really just want me to go 0+1 =1 which is odd then the other ones also?
that doesnt seem like the objective
1 and 2 are correct, what did I do wrong with 3?
I accidentally went -3 when deciphering and the first 2 ended up being correct, I don't understand...
Ah alright, it's decryption. That's why it's back 3
anyone has worked on game theory problems?
more exactly , collaborative game theory
look up evolutoinary biology
What approach can I use ?
counterexample
Would x = cuberoot(a) work?
what is a
a = 2 my b
@coarse gyro okay, can you show that x^2 is irrational and x^3 is rational?
if so you are done
Ah okay! Thanks @reef thistle
@reef thistle yo, I was trying to solve this problem: https://codeforces.com/contest/1209/problem/D
and so far you have?
Actually, I am new to graphs so I am kind of learning from the editorials
For the given input, I have drawn the graph
what's the graph?
okay so what do you notice?
edges are the animals so, I have joined the edges with their favorites
so how do you calculate minimum number of sad
If I give one edge between 3-4 their favorite then I will need to assign the other edge to any other node since that is not possible, at least 1 member is sad
because one edge will finish both of the nodes
Wait, if that is the case then only one will be happy
Say I have a K_n, how many will be happy?
K_n is?
complete graph, n vertices
does the answer change depending on what type of graph it is?
So, for n=4, it is 3, so n-1 is the maximum number for each n
woah
It doesn't change even when the graph changes
Everyone would be happy in this case that is n-1
@reef thistle You're a genius!
But, how was the M-(N-C) formula derived?
C is the number of connected components
M is 5
N is 6
C is 5
5-(6-5) = 4 that is not true
There are 5 connected components in the above graph, right?
in what graph
the chain you showed up there?
the one that has 1 connected to 2, 2 to 3, 3 to 4, 4 to 5 and 5 to 6?
this has only 1 connected component.
the graph is connected
no matter what two nodes you pick there's going to be a path between them
what led you to think there were 5 connected components?
Because 1 is connected to 2 so I thought that is also a component
no, {1, 2} is not a connected component.
because it's not maximal.
you can add more nodes to that without making it disconnected.
also, that ^ is a multigraph but it is indeed still connected so yes one connected component
and what do you mean by "complete these topics"
i'm not exactly sure what that'd entail
care to clarify?
Like what resource
Haha, no problem. Thank you for helping out!
i mean like... these were mentioned in my CS class back in second semester? and i had come across the terms on my own time before that? but i
can't really pinpoint a resource
You're another genius.
i disagree but thank you
Good night!
Hey :), I have to prove this equality
i did some basic steps
and now im thinking about using this formula
but its a lot of work
is there a simpler way to do this?
what do you mean by equality?
there's big O on both sides lol
is it an equality of sets of functions?
If I choose to prove the affirmative of a universal statement indirectly by way of contraposition, and during the proof of contraposition I encounter a contradiction , does this mean I proved the original statement?
hey can anypme help me with this question
hold up I will take a pic
this one thanmks
I know I have to use the binomial theorum
I just dont know what do to when the x and Y have coefficients
hey I tried the binomial theorum I think i did it incorectly but not sure what to do
cant I do 2^12 for this ?
list the possible things you can have
for variables, not the constants
you can have x^12, or x^11y, or x^10y^2, etc all the way to y^12
can you use that to figure out how many terms there can be
@reef thistle Why do we need to perform BFS/DFS for that question?
We just need to know the number of connected components, that's it?
We can check the adjacency matrix for the node that has whole column 0
That means that the node is connected to no one
that just tells you about isolated nodes, not connected components?
what question
@stray reef https://codeforces.com/problemset/problem/1209/D
You mean, have been discussed before?^
that just tells you about isolated nodes, not connected components?
Aah
one of them is an isolated node though
indeed
you guys are lovely!
start a dfs or bfs at a node
record every visited node
when the dfs or bfs ends, go to the next node
keep going to the next node until you reach one that hasn't been already been marked as visited (if you visited all nodes, then you're done)
restart at the first step
this algorithm is a pretty easy way to divide up the nodes into connected components
each bfs/dfs you do is one connected component
How do we ensure that we're not recounting a connected component? @lilac pivot as we're creating visited node each time?
because we're marking nodes as visited
in the second step
we never do a dfs on a connected component ever again
since it's already been marked as visited
as in the 2nd last line
when we do restart?
when we find a node that hasn't been visited
wow
let's take a look at this
we start at A, and find a search A,D,E,F
then we end the dfs
then we go to the next element
which is B
that hasn't been visited
so we start another dfs
and get B,C,H
then we go to C
that's been marked so we skip
then we skip D
then we skip E
then we skip F
then we get G, and we get a dfs, G
then we get H
and we skip it
so we're done, we did every node
we are left with 3 things:
A,D,E,F
B,C,H
G
which are obviously the 3 connected compenents
wow man!
This was DFS, let me try if BFS also checks the components
A -> D - > F
D - > F -> E
F->E
E
ending with ADFE marked as visited
bfs(B)
B->C
C->H
H
ending with BCH marked as visited
now only the node that is not visited yet is G therefore,
bfs(G)
G
ending with G marked as visited
No unvisited nodes left, total components = number of times BFS called?
Thank you!
Applications of BFS and DFS looks tougher ;-;
What’s the method of finding X in this kind of tasks. Substituting each of the possible answers takes too long, so I was wondering if there is another technique to approach that one
Can someone please help me understand what is expected of me here:
2. Fill in the blanks in the following sentence: If A, B and C are any sets, then by definition of
set difference x є A - (B ∩ C) if, and only if, x ________ and x ________ .
Would be that X is in the set of A and is not in the set of B ^ A or am I way off?
What do you mean by B ^ A
B and A.
How do you take the "and" of two sets?
Okay, so what's your final answer
If A, B and C are any sets, then by definition of
set difference x є A - (B ∩ C) if, and only if, x є A and x ∉ (B ∩ C).
But that seemed redundant. I felt like there had to be more
No that's it
Thank you!
- Prove or disprove the following statement by finding a counterexample.
For all sets A, B, and C, A ∪ (B ∩ C) is a subset of (A ∪ B).
Last Question. How can I use a counter example for proof? That would be endlessly exhaustive because it is always true.
Everything in A will always be apart of the set for both, and for set B, it will always have equal (if C has the same elements) or less (If C has some)
This statement says that it holds for all sets
So for a counter example, you just need to give an example of some sets where it doesn't hold
Since that would mean the statement is false since it claimed to work for all sets
I can not think of a set that it is not true for. As I explained before, A will always be apart of both A ∪ (B ∩ C) and (A ∪ B), and (B ∩ C) means that there will not be any elements that do not exist in B. Would it be false if it is equivalent? That is the only thing I can think. if B = C
I mean then maybe it's true? If you can't think of a counterexample
But if we are told "Find a counterexample" how would that work. if they are all true, wouldn't that be exhaustive checking and that goes infinitely.
But that's what proofs are for
@cerulean ore sounds like it, do you have the idea?
I am sorry. I am clearly misunderstanding something. If to get the question correct I need to find a counterexample. but a A counterexample is used to show a general statement is false. How can I get a counterexample if it is always true? Am I missing an option that is false?
Ah you might be reading the phrasing wrong here
Your two options here are
- Prove
or 2) disprove by finding a counterexample
You don't prove it by finding a counterexample
Ah. Okay thank you that makes sense.
So:
Let, x є A ∪ (B ∩ C)
x є A or B ∩ C
x є A or B
x є (A u B)
Yep, looks good
Hey can someone help me with the last part of this problem
guys
how I do this sum
$\binom{1000}{50} +2\binom{999}{49} + 3\binom{998}{48}.... .51 \binom{950}{0} $
epelta:
any halp?
I'd try rewriting it in terms of an index, then start to compare to things I already knew which were similar maybe, no guarantee
my guess it's something like the derivative of (x+y)^1000 with numbers plugged in
might have to do something to fix up stuff by playing around
$\binom{1000}{950} +2\binom{999}{950} + 3\binom{998}{950}.... .51 \binom{950}{950}$
epelta:
is the first one or the second one the correct question
wdym, I believe that they are different?
Which identity
epelta:
they are indeed same
yeah, my bad, sorry bout that
@weary tiger A lot of these types questions can be solved using a combinatorial argument. But since you mentioned hockey stick, try $\sum_{k=0}^{50}(k+1)\binom{1000-k}{950}$ and making a substitution of $ m = 1000-k$
Kei:
What's the method for determining if a linear recurrence is homogenous?
wait nvm, found it in my notes
@reef thistle As I am learning, I just checked the editorial
it was a bipartite problem
Hey, I’m new to this course and for my assignment I have to find the complement of :
(A-(A u B)) U (A-(B-A))
This whole equation is just equal to A (If I’m not wrong) so would the complement of it just be Ā ?
indeed
Thanks!!
can anyone explain how y=x^3 is an one to one function or injective?
y(1) = 1, y(-1) = -1
this contradicts the definition of one to one in my notes. one to one must have different outputs and inputs.
i was also given that it has to pass the vertical line test, but im confused which is correct.
by different outputs and inputs that means something else than that
for instance f(x)=x^2 this would not work for f(1)=f(-1) = 1
that's because f(1)=1 and f(-1)=1, they have the same output, so it should be the same input, but the inputs are different
you can see why this is failing the vertical line test because it cuts the parabola at both of those points
this contradicts the definition of one to one in my notes. one to one must have different outputs and inputs.
how does it
which makes it one to one.
1 is not the same as -1
I see how it can be interpreted wrong, "have different outputs and inputs" does sound like f(a)=a is not allowed because the output and input are the same
@lethal sequoia COuld you elaborate on what you mean by combinatorial arguements?L
$\binom{m}{k} = \binom{m}{m-k}$ right?
epelta:
yes
Yes but im not sure if standart method works with inequalities as well
what do you mean by "standard method"
First you check if it works for 0 in this case 2
Then you write it for m and after that (m+1)
the format of a proof by induction is the same regardless of the nature of the statement being proved. why would it be any different if the statement to be proved happens to be an inequality?
Yes i get that
But im not sure about next steps
This is what im writong and im not sure if its correct: "2-(1/m)+(1/(m+1)²) <=2-(1/(m+1)
Wait
I think i got it
No worries
36 out of what
36
nice
Thank you guys for helping me! You are my real teachers.
@reef thistle @stray reef @pale epoch and many others!
Uhm, is 63^1337 mod(64) the same as (63 mod(64)^1337?
hmm
yep
the parens don't match though
63 mod (64) = 63 no?
well, there's another number it is congruent to
that is easier to carry calculations with
maybe, i'm just wondering why i'm getting the second one to 63^1337 and the first one should be 63 according to my calculator 😄
yeah, it is 63
Nvm I think i just got a little confused with equal/congruent ^^
So how would I go finding that easier number that's congruent @reef thistle were talking about?
nice
Hey guys, just a quick question...
- Determine whether each of these functions from Z to Z is one-to-one.
a) f(n)=n2 +1
a) f(n) = n3
When given a question like this I know how to solve but is it possible that you could have a question in which instead of Z to Z, it was something different (e.g R to Z) ?
Yes
They can change the domain/co-domain and ask if still the function is one-one or onto (or both)
So for that question if we had f(n)=x^3 for aEZ and bER
How would it change the way we solve it?
Changing the codomain doesn't really how you solve the one to one part
But it does change the onto part
But, changing the domain changes how you do both parts
Ohh okay thank you so much!
Yeah okay so I figured out that 63 would be congruent to -1, which made everything a little bit easier
i tried they dont seem to work
i tried perfect squares
idk if theres another way of doing it other than the method of exhaustion
You can't try every real number lol
There's a very simple solution here. If you're thinking squares, you're already overthinking it
The solution is a = 0, b = 0
When finding examples/counter examples, try to think of 0 objects first as they're usually the right guess
idk from what i learned say u try to express 2 different even numbers u have to use different variables
a=2k b= 2j
or something like that
Seems like you've been working with vectors, but these are real numbers
vectors?
The question is looking for 0,0 and might not like any other format
0
oh
okok
but say if i run across a problem with different variables , can they have the same value?
Yes of course. They don't have to be different unless the question makes it clear they are supposed to be
"distinct" is the word commonly used to say two variables have to have different values
oh ok
i dont get why 3 is not right :/
2(3)^2 -5(3) +2 = 5
and 5 is a prime number
oh wait i had to put in the 5 too
smh
~-~
Hey, I have function f(h)=f(n)+O(g(n)) where g(n)=o(f(n)) how can i try to estimate 1/h(n)? D:
@weary tiger With a combinatorial argument, try to make an argument by counting instead of using algebra
@lethal sequoia what do you mean?
I wanted to see which values are both in the fibonacci sequence and sum from 0 to n sequence, and tried programmatically with values close to long overflows, and didn't found any match except the 4 first, is there a mathematical way to prove that there is no other match except those 4 first, even going up to infinity?
hmm
@earnest canopy Well I know the only powers in fibonacci are 8 and 144...
there's a good chance we can prove
so for the sum 1 to n sequence, k is a sum from 1 to n if (and only if) 8k+1 is a square.
A number k is Fibonacci iff at least one of (5k^2 + 4) or (5k^2 – 4) is a square
Let's just whack 5((a^2+a)/2)^2+4 = b^2
5(a^2+a)^2+16 = 4b^2
5(a^2+a)^2 = 4(b+2)(b-2)
gcd(b+2, b-2) is either 1, 2, 4.
if 1, then either (b+2)=5c^2, (b-2)=d^2 or (b+2)=c^2, (b-2)=5d^2, probably can attack from there.
so 5c^2-4 is square in first case, 5d^2-4 is square in second case... whoa new fibonacci numbers
I read everything, but my brain isn't big enough to process everything you wrote I guess 😂
but didn't know about those properties of "8k+1 have to be a square to be a 1 to n sum", pretty interesting
so here you brought some paths but no conclusion? I don't think I'm good enough to help you continue the path 😦
also what is gcd function?
gcd is greatest common divisors
ok, know those names in french but not in english, thx 👌
What have you tried?
Idk where to start so no
Think about it
Do I need to prove both big Oh and omega?
But how to represent the precision smybol
Element118:
@raven lodge
Hmm can I directly prove big theta taking log in both sides?
the first line you would need to be careful by bounding the proportionality
I have to write down the equation is larger than 1 and smaller than 2?
Or do I need to explain more on why 1+e =2^e?
you probably should write that n^(2^-k)<1+e<n^(2^-(k-1))
Okay thanks
Need help with the following probability question: If there is a best of 7 soccer series, what is the probability that the team that wins the first game goes on to win the series. Assume that both teams are equally likely to win the game and the games are independent events.
Casework: what’s the probability that the team wins in 4,5,6,7 games
@sweet island Would that be (1/2)^6? Because there are 6 more series and in each series it's a coin toss between the teams
The answer is 21/32
How does the matching of team works?^
Category wise?
one team vs all other teams?
oh
My answer is in bold. I am struggling a bit with understanding. I have watched a few videos and read the book. My understanding is it may be possible with a subcircuit? If anyone has a resource to help me understand please link me.
@outer spire If you don't mind me asking, why?
@clever nacelle i doubt you defined eulerian circuit that way, usually this is a theorem, but that's correct so far
for the other question your idea is correct
but it could be explained better
you can maybe split it into 3 cases: starting in the "left" part of the graph, the "right" part and starting at i
that would do
basically
for hamiltonian cycle, in general, if you remove k vertices, you should be left with at most k connected components
Hey guys, just joined the discord. I'm working on some exercises from Rosen's discrete math book and am stuck on a question. How is 8x2^n-1 turning into 8x2x2^n-1.. where is the second 2 coming from?
sorry that's what I meant 8x2x2^n-2
I just did some research, and it's an exponent rule that lets us do that
definitely gotta brush up on my fundamentals, it's been a while since I was in school and am struggling in college
can someone possibly help me wif this xd
@ocean viper simulate the problem
im not sure how to approach simulating it
at 8 min thaat guy will be at the start and the 10min guy will be still running
at 10 min the 8min guy will be running? aanad the 10 min guy aat starting?
8,16,24,32,40
10,20,30,40
so...?
More generally, if n runners are simultaneously running with constant speeds on a circular track, beginning at the same spot at the same time, and for all j, the jth runner first returns after aⱼ minutes, then the jth runner will be at the start after i minutes if and only if j divides i, and thus the least number of minutes after which all runners will be at the start simultaneously is the least common divisor of all the aⱼ's, that is lcm(a₁, a₂, ..., aₙ).
your solution is correct, I just proved a more general statement
It is Cartesian product of sets
so {A U B} , {A U C}?
no
the set of ordered pairs whose first element is in A and whose second element is in B
$A \times B$ consists of ordered pairs $(a,b)$ where $a \in A$ and $b \in B$
Ann:
So, for example (1, green) is a member of Integers * Colors
for example
if you take S = {spades, hearts, diamonds, clubs} and R = {A, K, Q, J, 10, 9, ..., 3, 2}
then S x R is a deck of cards
oh wait i remember now
say a = {1,2,3} b {4,5,6}
axb would be {1,4} {1,5} , {1,6} {2,4} {2,5} {2,6}??
and so on
no
o-o
It would be {(1,4),(1,5),(1,6),...}
o
okok
not sure if this is the right channel for this, since this is a question from one of my discrete math courses, but i was wondering if i could get some help in solving this one?
i'm torn between k^n and n!k
i have a feeling it might have something to do with n choose k or stirling numbers of the second kind
Let's say you have k empty boxes. And you have n objects that you can put in them
How would you approach this problem?
@wanton sable
i figure it would have something to do with factorials right? like n choices for box k, n-1 choices for box k-1, etc
or is that completely off?
k ways?
n!k ?
No!
Each object can be put in k ways
And you have n such objects
Yes, it is k^n
But did you really understand?
You're welcome then!
Hi
Im currently studying DFA/NFA
Im struggling with the language here
why is there a hat on the transition function
Could somebody put into english what that is trying to say? Thanks 🙂
I’ve just now heard of DFA so I can’t really help except to search stuff on google. Maybe this will help
@subtle halo forgot to ping you
@subtle halo yeah, it mostly likely refers to the extended transition function
to put it into english, the transition function takes a state and a single letter and outputs the state reached
the extended transition function can take in a state and a string (multiple letters) and outpute the state when entering this string
in this case i assume $\widehat{\delta}(x)$ is the state that is reached when inputting the string x
Lochverstärker:
yes, that's it
L(M) is the language accepted by the DFA, i.e. it's all strings that when entered end up in a final state
hi, does anyone know if there is any generalization of the knapsack problem with lower and upper bounds? So the elements that one can include in the knapsack such that the knapsack does not have a weight less than N and more than M ?
shouldn't be too hard to generalise
@pale epoch ah that makes sense, thank you!!
hi
can someone help me undertsand a few things about logic?
in discrete mathemetics terms
I know that P -> Q is synonymous with ~P v Q, but is ~P -> Q synonymous with just P v Q? Or do I follow demorgans laws here?
?



