Great Theoretical Ideas in Computer Science. Lecture 5: Cantor s Legacy

Size: px

Start display at page:

Download "Great Theoretical Ideas in Computer Science. Lecture 5: Cantor s Legacy"

Brandon Marshall
5 years ago
Views:

15-251 Great Theoretical Ideas in Computer

1 Great Theoretical Ideas in Computer Science Lecture 5: Cantor s Legacy September 15th, 2015

2 Poll Select the ones that apply to you: - I know what an uncountable set means. - I know Cantor s diagonalization argument. - I used to know what uncountable meant, I forgot. - I used to know the diagonalization argument, I forgot. - I ve never learned about uncountable sets. - I ve never learned about the diagonalization argument.

3 This Week All languages? Decidable languages Factoring Regular languages EvenLength. 0 n 1 n Primality.

4 Our heroes for this week father of set theory father of computer science Uncountability Uncomputability

5 Infinity in mathematics Pre-Cantor: Infinity is nothing more than a figure of speech which helps us talk about limits. The notion of a completed infinity doesn't belong in mathematics - Carl Friedrich Gauss Post-Cantor: Infinite sets are mathematical objects just like finite sets.

6 Some of Cantor s contributions > The study of infinite sets > Explicit definition and use of 1-to-1 correspondence - This is the right way to compare the cardinality of sets > There are different levels of infinity. - There are infinitely many infinities. > N < R even though they are both infinite. > N = Z even though N(Z. > The diagonal argument.

7 Reaction to Cantor s ideas at the time Most of the ideas of Cantorian set theory should be banished from mathematics once and for all! - Henri Poincaré

8 Reaction to Cantor s ideas at the time I don t know what predominates in Cantor s theory - philosophy or theology. - Leopold Kronecker

9 Reaction to Cantor s ideas at the time Scientific charlatan. - Leopold Kronecker

10 Reaction to Cantor s ideas at the time Corrupter of youth. - Leopold Kronecker

11 Reaction to Cantor s ideas at the time Wrong. - Ludwig Wittgenstein

12 Reaction to Cantor s ideas at the time Utter non-sense. - Ludwig Wittgenstein

13 Reaction to Cantor s ideas at the time Laughable. - Ludwig Wittgenstein

14 Reaction to Cantor s ideas at the time No one should expel us from the Paradise that Cantor has created. - David Hilbert

15 Reaction to Cantor s ideas at the time If one person can see it as a paradise, why should not another see it as a joke? - Ludwig Wittgenstein

16 First we start with finite sets

17 How do we count a finite set? A = {apple, orange, banana, melon} What does A =4 mean? There is a 1-to-1 correspondence (bijection) between A and {1, 2, 3, 4} apple orange banana melon

18 How do we compare the sizes of finite sets? A = {apple, orange, banana, melon} B = {200, 300, 400, 500} What does A = B mean? apple orange banana melon

19 How do we compare the sizes of finite sets? A = {apple, orange, banana, melon} B = {200, 300, 400, 500} What does apple orange banana melon A = B mean? A = B iff there is a 1-to-1 correspondence (bijection) between A and B.

20 How do we compare the sizes of finite sets? A = {apple, orange, banana} B = {200, 300, 400, 500} What does A apple B mean? apple orange banana

21 How do we compare the sizes of finite sets? A = {apple, orange, banana} B = {200, 300, 400, 500} What does apple orange banana A apple B mean? A apple B iff there is an injection from A to B.

22 How do we compare the sizes of finite sets? A = {apple, orange, banana} B = {200, 300, 400, 500} What does apple orange banana A apple B mean? A apple B iff there is a surjection from B to A.

23 3 important types of functions injective, 1-to-1 f : A! B is injective if A,! B A B a 6= a 0 =) f(a) 6= f(a 0 ) surjective, onto f : A! B is surjective if 8b 2 B,9a 2 A s.t. f(a) =b A B A B bijective, 1-to-1 correspondence f : A! B is bijective if A $ B f is injective and surjective A B

24 Comparing the cardinality of finite sets A B A apple B A,! B A B A B A B A B A = B A $ B

25 Sanity checks A apple B i B A A,! B i B A A = B i A apple B and A B A $ B i A,! B and A B A $ B i A,! B and B,! A If A apple B and B apple C then A apple C If A,! B and B,! C then A,! C

26 One more definition A < B not A B There is no surjection from A to B. There is no injection from B to A. There is an injection from A to B, but there is no bijection between A and B.

27 So what is the big deal? This way of comparing the size of sets generalizes to infinite sets! These are the right definitions for infinite sets as well!

28 Comparing the cardinality of infinite sets A apple B A,! B A B A B A = B A $ B

29 Sanity checks for infinite sets A apple B i B A A,! B i B A A = B i A apple B and B apple A A $ B i A,! B and A B A $ B i A,! B and B,! A Cantor Schröder Bernstein If A apple B and B apple C then A apple C If A,! B and B,! C then A,! C

30 So what is the big deal? Let me show you some interesting consequences.

31 Examples of equal size sets N = Z N = {0, 1, 2, 3, 4,...} Z = {..., 4, 3, 2, 1, 0, 1, 2, 3, 4,...} , 1, 1, 2, 2, 3, 3, 4, 4,... f(n) =( 1) n+1 l n 2 m

32 Examples of equal size sets N = Z Does this make any sense? A ( B =) A < B? N(Z Shouldn t N < Z? Does renaming the elements of a set change its size? No. Z Let s rename the elements of : {...,banana, apple, melon, orange, mango,...} Let s call this set F. How can you justify saying N < F? Bijection is nothing more than renaming.

33 Examples of equal size sets N = {0, 1, 2, 3, 4,...} S = {0, 1, 4, 9, 16,...} N = S f(n) =n 2

34 Examples of equal size sets N = {0, 1, 2, 3, 4,...} P = {2, 3, 5, 7, 11,...} N = P f(n) =n th prime number.

35 Definition: countable and uncountable sets Definition: A set A is called countable if A apple N. A set A is called countably infinite if it is infinite and countable. A set A is called uncountable if it is not countable. (so A > N )

36 How to think about countable sets A set A is called countable if A apple N. So why is it called countable? A apple N means there is an injection f : A! N. you could count the elements of A (but could go on forever) A N

37 How to think about countable sets A set is called countable if. A A apple N Perhaps a better name would have been listable: can list the elements of A so that every element appears in the list eventually. a 1 a 2 a 3 a 4 a 5 (this is equivalent to being countable)

38 How to think about countable sets A set A is called countable if A apple N. This seems to imply that if A is infinite, then A = N. Is it possible that A is infinite, but A < N? Theorem: A set A is countably infinite if and only if A = N. So if A is countable, there are two options: 1. A is finite 2. A = N Exercise: prove the theorem

39 Countable? N = Z Z?. (0, 0) (1, 0) (1, 1) (0, 1) ( 1, 1) ( 1, 0) ( 1, 1) (0, 1) (0, 0) (1, 1) (2, 1) (2, 0) (2, 1) (2, 2) (1, 2). (0, 2).

40 Countable? N = Q? Can we list them in the order they appear on the line? Between any two rational numbers, there is another one. a Any rational number can be written as a fraction. a b Z Z Q ( map (a, b) to ) b =) Q apple Z Z = N

41 Countable? N = {0, 1}? {0, 1} = the set of finite length binary strings. " , 01, 10, , 001, 010, 011, 100, 101, 110, 111

42 Countable? N =? = the set of finite length words over. Same idea.

43 Countable? N = Q[x]? Q[x] Take = the set of polynomials with rational coefficients. e.g. x x2 +6x ={0, 1,...,9,x,+, 22 7,, /,ˆ} Every polynomial can be described by a finite string over. e.g. xˆ3 1/4xˆ2 + 6x 22/7 So Q[x] i.e. Q[x] apple

44 The CS method for showing a set is countable CS method to show a set A is countable ( A apple N ) : Show A apple for some alphabet. i.e. A i.e. Show that you can encode the elements of using finite length words over an alphabet. A

45 Seems like every set is countable Nope!

46 Cantor s Theorem Theorem: For any non-empty set A, A < P(A). S = {1, 2, 3} P(S) ={;, {1}, {2}, {3}, {1, 2}, {2, 3}, {1, 3}, {1, 2, 3}} P(S) = 2 S P(S) ${0, 1} S S = {1, 2, 3} binary strings of length S 1 0 1! {1, 3} 0 0 0! ;

47 Cantor s Theorem Theorem: For any non-empty set A, A < P(A). So: N < P(N). I.e. P(N) is uncountable. N < P(N) < P(P(N)) < P(P(P(N))) < (an infinity of infinities)

48 Cantor s Theorem - Proof by diagonalization Assume P(A) apple A for some set A. So A P(A). Let f be such a surjection. Example 1 N P(N) {3, 7, 9} {2, 5} {1, 2, 3} Define S = {a 2 A : a/2 f(a)} 2P(A). Since f is a surjection, 9s 2 A s.t. f(s) =S. But this leads to a contradiction: if s 2 S then s 62 f(s) =S if s/2 S then s 2 f(s) =S s S S = {1, 4,...} Is s 2 S? Why is this called a diagonalization argument?

49 Cantor s Theorem - Proof by diagonalization Example 1 {3, 7, 9} N P(N) 2 3 {2, 5} {1, 2, 3} S = {1, 4,...} 4... f(1) f(2) f(3) f(4) f(5). f(s)= S S is defined so that S cannot equal any f(i)

50 Let {0, 1} 1 Uncountable sets be the set of binary strings of infinite length. {0,1,2,3,4,5,6,7,8,9, } ! ;! N! {even natural numbers} {0, 1} 1 is uncountable, i.e. {0, 1} 1 > N because {0, 1} 1 $P(N). (just like {0, 1} S $P(S) ) (Recall {0, 1} is countable.)

51 Let {0, 1} Uncountable sets be the set of binary strings of infinite length. {0, 1} 1 is uncountable, i.e. {0, 1} 1 > N Direct diagonal proof: Suppose {0, 1} 1 apple N N {0, 1} 1 > cannot appear in the list

52 Uncountable sets R is uncountable. In fact (0, 1) is uncountable. exercise Be careful: =

53 Appreciating the diagonalization argument If you want to appreciate something, try to break it Exercise: Why doesn t the diagonalization argument work for N, {0, 1}, a countable subset of {0, 1} 1?

54 Uncountable sets Let be the set of bijections from to. B N N B is uncountable. CS method to show a set A is uncountable ( A > N ) : Show A {0, 1} 1 i.e. A {0, 1} 1 i.e. Show that the elements of A encode all the elements of {0, 1} 1.

55 One slide guide to countability questions You are given a set A. Is it countable or uncountable? A apple N or A > N? A apple N : - show directly that A,! N or - show A apple B, where A > N : N A B 2{Z, Z Z, Q,, Q[x]} - show directly using a diagonalization argument - show A {0, 1} 1

56 An Interesting Question Is there a set S such that N < S < P(N)? Continuum Hypothesis: No such set exists. (Hilbert s 1st problem)

57 The story continues next lecture

MATH 3300 Test 1. Name: Student Id:

MATH 3300 Test 1. Name: Student Id: Name: Student Id: There are nine problems (check that you have 9 pages). Solutions are expected to be short. In the case of proofs, one or two short paragraphs should be the average length. Write your