Skip to main content
OlympiadHQ

Browse · MathNet

Print

Selected Problems from Open Contests

Estonia counting and probability

Problem

Prove that the set of integers can be partitioned into disjoint subsets such that both of the following hold:

a) If , then the subsets and have the same number of elements.

b) If and are non-negative integers and , then for an arbitrary element in the set , there exist elements and from the sets and , respectively, such that .
Solution
Divide the set into subsets such that the subset consists of only those numbers which have exactly ones in their binary representation. Then . Let us show that both conditions are met.

The first condition is met because the numbers with ones are in one-to-one correspondence with the numbers with ones: given a number, simply replace all ones in its binary representation by zeros and vice versa.

To show that the second condition is met, choose an arbitrary number from the set . Its binary representation contains exactly ones. Construct a binary number by choosing ones from the binary representation of and filling all other binary places by zeros, analogously construct a second number based on remaining ones in . Then and .

---

Alternative solution.

Let us prove the statement by induction. If , then , and taking and we get a partition that satisfies both of the requirements.

Assume now that we have a partition for the set . Construct a partition of the set based on that. First, generate the sets as follows: the elements of the subset are derived from the elements of the subset by adding to them. The subsets form a partition of the set and from the construction for all the corresponding subsets and have the same number of elements.

Now, let for all and in addition to that, and . Then, , and if , then also , and . As , we see that and ; thus .

To verify that the second condition is met, let be an arbitrary element of . If , then and , so we can take and . Now assume . If , then is an element of and thus there exist elements and in the sets and , respectively, such that . If , then is an element of , i.e. is an element of the set and thus the sets and contain elements and , respectively, such that . But now , where and are elements of the sets and , respectively. So the statement holds for all positive .

Techniques

Recursion, bijectionAlgebraic properties of binomial coefficientsInduction / smoothing