Kneser's conjecture: the minimum number of colors in order to color the K(n,k) is n-2k+2. Where K(n,k) is the graph on "n choose k" vertices were two vertices are connected iff they are disjoint.
Proof: assume by way of contradiction that there is a coloring in n-2k+1 colors. Define d=n-2k+1 and consider the d-dimensional sphere. Place n points numbered from 1..n on the sphere such that no d+1 points are on the same "equator".
Next, define open sets C_1,...,C_d as follows: C_i = {x: the half sphere were x is the center contains k points and the coresponding vertix in G is colored i}. Define F = S^d \ Union(C_i).
By Bursuk-Ulam for every cover of the n-dimensional ball (using l1 norm) using n+1
open/closed sets there is a point x for which x,-x are in the same set.
We get that there is a set with x,-x in it:
- If it's F, we reach a contradiction because there are more that d+1 vertices on the equator between x and -x.
- If it's one of the C_i-s we reach a contradiction: because the points that x "sees" are disjoint from the ones -x "sees". So in K(n,k) there is an edge between them - thus the coloring is not valid.
QED
Prove that if it were true then there must be a contradiction.
prove it
scientific
Give a counter-example.
Yes.
by bringing evidence to the table
A counter example
Which part of a 2-column proof is the hypothesis of the conjecture? GIVEN
Find one counterexample to negate the statement
One is enough.
Nobody has been able to prove it so far, so you'll have to come up with some very original idea if you want to prove it.
It does not necessarily prove congruence but it does prove similarity. You can have a smaller or bigger triangle that has the same interior angles.