Olympiad Maths Prep

Track / Stage 6 / 34 of 400 #1034 of 2000

Problem 1034

National olympiad, first round
Number theory Difficulty 6.0 Prove it

Let's prove that no matter how we divide the natural numbers into two subsets, there will always be at least one subset containing two numbers whose arithmetic mean is an integer, and both numbers are in the same subset.

This one wants a proof. Work it on paper, read the official solution, then mark yourself honestly — the ladder only means something if the record is true.

Official solution

Let the two subsets be denoted by AA and BB, and that nn is an element of AA is denoted (as usual) by nAn \in A. We assume that the statement of the problem is not true, and A,BA, B is such a partition where there is no three-term arithmetic progression. We will show that if n4n \geq 4, then (n1)(n-1) and (n+1)(n+1) cannot belong to the same set. If indeed n1A,n+1An-1 \in A, n+1 \in A, then only n3B,n+3Bn-3 \in B, n+3 \in B could be possible, but then nn could not be in either AA or BB. Clearly, we can assume that 7 is in AA. Then, according to our previous finding, only 5B,9B5 \in B, 9 \in B can be true, and because of these, 3A,11A3 \in A, 11 \in A. Since the arithmetic mean of 3 and 11 must be in AA according to our assumption, we have reached a contradiction, so our statement is true.

Remark. It can also be shown that the first 9 natural numbers cannot be partitioned into two sets such that neither contains a three-term arithmetic progression. This is not true for the first 8 numbers, as the partition A={1,2,5,6},B={3,4,7,8}A=\{1,2,5,6\}, B=\{3,4,7,8\} demonstrates. As can be inferred from the theorem recently proved by mathematician Endre Szemerédi, if we partition a sufficiently large set of numbers into two sets, the existence of arithmetic progressions of arbitrary length can also be proven.

Source: NuminaMath-1.5, licensed Apache-2.0. Statement and solution reproduced as published; topic, difficulty and ordering added by this site.