A subset of does not contain three elements whose product is a perfect square. Determine the maximum number of elements in .
Solution
By checking the parity of the prime factorization of all the integers in , we partition them into disjoint triples, whose products are all squares:
If , at least one of these triples will be in , a contradiction.
We can easily check that the subset (obtained by strategically removing one element , , , , from each of the triples above):
satisfies the condition given. Thus the maximum number of elements in is .
Want a route through all this instead of an archive? The track
puts 2,000 problems in a working order, from AMC 10 level to the IMO shortlist.