Combinatorics - Chapter 1 Flashcards

1
Q

pigeonhole principle

A

Let k∈ℕ be the number of colours and s₁,..,sₖ be non -ve integers . Let [n] be coloured with colours c₁,…,cₖ and n> the sum of the sᵢs. Then there exists iϵ[k] s.t. there are at least sᵢ+1 numbers with colour cᵢ.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

s₁,..,sₖ in pigeonhole principle

A

the sizes of each hole.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

1,…,n in pigeonhole principle

A

the pigeons

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

c₁,…,cₖ in pigeonhole principle (colours)

A

the pigeonholes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

ramsey number

A

For s,t∈ℕ the ramsey number of s and t denoted by R(s,t) is the smallest n s.t. every red/blue-edge-colouring of Kₙ contains a red Kₛ or a blue Kₜ.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

ramsey number informal

A

the smallest n s.t. no matter how you colour the edges of Kₙ with red and blue you can always find a red Kₛ or a blue Kₜ.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

R(t) =

A

R(t,t)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

How to show that R(s,t)=n

A
  • LOWER BOUND show R(s,t)>n-1 via an explicit counter example
  • UPPER BOUND show Kₙ contains a red Kₛ or a blue Kₜ
  • upper and lower bound together makes R(s,t)=n
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

R₁(s₁)=

A

s₁ because every colouring of Kₙ with one colour contains a monochromatic Kₛ₁ (if s₁<=n)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

R₂(s₁,s₂) =

A

R(s₁,s₂) (i.e. the original ramsey definition)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

R(3,3)=

A

6

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

R(3,4)=

A

9

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

R(3,5)=

A

14

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

R(4,4)=

A

18

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

R(4,5)=

A

25

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

R(5,5)=

A

in the range 43-48

17
Q

How many ways are there to red/blue colour Kₙ

A

2^ (n choose 2) ways

18
Q

How many edges in a complete graph

A

n choose 2

19
Q

What is the bound for R(t)

A

( square root 2)ᵗ < R(t) ⩽ 4ᵗ

20
Q

R(Kₛ,Kₜ)=

A

R(s,t)

21
Q

Van der wearden numnber

A

For any r,m∈ℕ the van der weerden number of r and m denoted by W(r,m) is the smallest n such that whenever [n] is coloured with r colours, then there exists a monochromatic arithmetic progression of length m.

22
Q

n choose k

A

n! / k!(n-k)!

23
Q

binomial distrubution

A

(n choose x)(p)^x(1-p)^(n-x)

24
Q

erdos and szekeres

A

R(s,t)⩽R(s-1,t)+R(s,t-1)

25
Q

number of edges in a complete graph Kn

A

n choose 2 = n(n-1)/2

26
Q

pigeonhole principle simple

A

If at least n pigeons are placed into k pigeonholes then some hole contains at least the ceiling of n/k pigeons.

27
Q

First step when proving R(s,t)⩽(an expression)

A

let n=expression.

28
Q

How to show R(s,t)⩽n

A

Show any red/blue-edge-colouring of Kₙ contains a red Kₛ or a blue Kₜ.

29
Q

How to show R(s,t)≥n

A

Show R(s,t)>n-1. Find an explicit edge colouring without a red Kₛ or a blue Kₜ.

30
Q

(n choose t) is the number of ways….

A

to choose t vertices in Kₙ

31
Q

Expectation of X =

A

sum over i in natural numbers of i multiplied by the probability X=i

32
Q

colour blindness argument

A
  • Define the colouring with the bigger amount of colours
  • From this colouring define another colouring on a smaller amount of colours where you colour with a certain colour if it was coloured with a group of colours in the original colouring.
  • Apply Ramsey definition and look as cases.
33
Q

when to use a colour blindness argument

A

when showing inequalities of Ramsey numbers with different numbers of colours

34
Q

Erdos and Szekeres informal

A

R(s,t)⩽R(s-1,t)+R(s,t-1)

35
Q

Add probabilities if

A

one event or the other

36
Q

Multiply probabilities if

A

probabilities both events will happen

37
Q

Conditional probability

A

P(A|B)=P(A∩B)/P(B)

38
Q

Erdos Theorem Quick

A

Let t,n∈ℕ. If (n chose t) * 2^(1- (t choose 2))<1, then R(t)>n.

39
Q

Schurs Theorem

A

For k∈ℕ there exists n=n(k)∈ℕ such that every k colouring of [n] there exists integers x,y,z∈ℕ all coloured with the same colour s.t. x+y=z