Cantor's theorem : show that there exists no surjective function from a set to its powerset
i don't get the proof, looking for somebody to discuss with
this proves that the cardinality of the power set X has cardinality greater than the cardinality of set X
| P ( X ) | > | X |
i would have to look at the proof
yes i have the video if u have time
the video?
ok
fastforward to 25th minute
where is the video?
its a bit hard to see, the graphics is a bit blurry
watching :)
yeah quality isn't that good but he explains quite well, however im not getting the how the new subset won't exist in any of the previous subsets
suppose there exists a bijection between A and 2^A
k
2^A is a shorthand way of saying the powerset of A
okay
sorry, im watching , i will respond soon
it is easy to see for finite sets because n < 2^n, i have trouble with understanding the argument for infinite sets
sure take your time :)
his pictures are a bit confusing
ok lets say A = { S, C , T } , which stands for square circle triangle
lets construct P(A) = { {S} , { C }, { T}, { S, C }, { S, T } , { C , T } , { S , C , T } }
and i will define a function if 'a' is an element of A , f(a) is a subset of A
ok
would it be general enough for infinite sets like A = all natural numbers ?
yes, you can generalize this
its easier to show with finite sets, and then you get the idea
okay
so i will show that f is not onto. i will construct a subset of A for which there is no element in the domain of f that maps to it
since the codomain of f are subsets of A
I see, constructing a subset that cannot be an image of an element in domain will do
{S, C, T} ----> { {S} , { C }, { T}, { S, C }, { S, T } , { C , T } , { S , C , T } }
so define this subset of A call it B, B = { a | a is not a member of f(a) }
could you please elaborate, this is the part thats been tripping me
there is more than one possible function this is one possibility S -> {S} C -> { T} T -> { S, C, T }
we are mapping elements of A to subsets of A
also you need to include the empty set in the power set of A (just a small change there)
yes what do they mean by "a is not a member of f(a)"
f(a) is a subset right ?
ok so we want to construct a set B , so in our case we would have
right f(a) is a subset of A
and the new set B is also a subset of A ?
yes,
what if S-->{S} C-->{C} T-->{T} ?
oh
oops i made a mistake
B cannot have S/C/T because they exist in f(S), f(C) and f(T) is it ?
if our function f is S -> {S} C -> { T} T -> { S, C, T } then B = { C}
I think im getting it
if our function f is S-->{S} C-->{C} T-->{T} then B = { }
sorry that B should just be the empty set
since S elt. of {S} , C elt. of {C} , and T elt. of {T } there are no elements of A which are not elements of f(A), so B is just the empty set
but notice that in that case, there is no element in A being mapped to the empty set, and the empty set is an element of P(A)
I see the logic now thanks :) how do I put it in steps for a contradiction proof i don't understand his steps
His proof starts with defining a new subset so define this subset of A call it B, B = { a | a is not a member of f(a) }
right
it assumes that B is already in image of f
and arrives at contradiction
no, we don't assume that
B is in P(A) , or the codomain of f , by definition
since it is a set of members of A, so it is a subset of A
well, i guess if you are doing a proof by contradiction, since you are assuming that it is bijective, then yes you are right
woops, sorry
ok then if there is a bijection, then yes B must be in the image of f. but this will lead to a contradiction
ok so is there an element a of A such that f(a) = B ?
yes so far im with him, his next steps are very confusing
since B is in range of f, there exists some "a" in A such that f(a) = B fine
how does this lead to contradiction
if there exists an element a such that f(a) = B, then either a is inside B or it is not inside B .
ok
if a belongs to B , then a does not belong to B. if a does not belong to B , then a belongs to B
B = { a | a is not a member of f(a) }
lets use their example
ok going thru pdf
there are a few obvious typos in the pdf, though, when you get to the cantor theorem section
Let S= { 1,2,3,4} Define a function f 1 -> { 1,3} 2 -> {1,3,4} 3 -> { } 4 -> {2,4} Let X = { s | s not elt. of f(s) } then X = { 2,3 } Now the question: Is it possible that there exists some s in S such that f(s) = X ? If there does exist some s such that f(s) = X , (which we need if f is onto) then it is a fact that either s is in X or s is not in X. If s is in X , then s is not in X. If s is not in X , then s is in X.
suppose there was an element such that f(s) = X f(s) = { 2,3 } , in our example
s -> { 2 , 3 }
ok X = {2, 3}
either s is inside { 2, 3 } or is not inside { 2 , 3 } case 1 : s is inside { 2, 3 }
Ohh thats not possible because by definition \(s \not \in f(s)\)
right
by definition of X
X contains all elements who map to subsets that dont contain the element itself
s ---> f(s) if f(s) = X, then X cannot contain s. clear
right, but then what happens when s is not in { 2, 3 }
so we established that s -> { s , 3 } s -> { 2, s } are impossible but so is s -> { 2, 3 } impossible,
where s =/= 2, and s =/= 3
I get that s cannot be in X why s->{2, 3} is impossible ?
lets look at the definition of X again
Let X = { s | s not elt. of f(s) }
Oh if it were possible to have a mapping such as : s->{2, 3}, the s must exist in X !!
if s -> { 2, 3 } and s =/= 2 and s =/= 3, then s must be inside X by definition
beautiful! i see how contradiction works, thanks a lot for explaining clearly and being patient :))
but then we get back to where we started
Join our real-time social learning platform and learn together with your friends!