Erdős Problem #1207 — Large subsets avoiding isosceles triangles
Let be positive integers, and let be the largest integer such that every set of exactly distinct points has a subset of size at least containing no three distinct points with . Thus an isosceles triple is forbidden even when its three points are collinear and equally spaced. The question is to estimate , especially whether there are constants and such that, for every integer ,
Erdős Problems #1207 The Minkowski grid has robustly many repeated distances
Background
Erdős’s 1980 survey attributes the question to Riddell and asks for estimates in general dimension as well as in the plane. The one-dimensional case is the problem of finding large subsets of a finite set of real numbers with no nontrivial three-term arithmetic progression: three distinct points on a line form an isosceles triple precisely when the middle point is the average of the other two. Erdős also suggested that lattice points in a sphere of suitable radius might provide particularly restrictive examples. A related basic observation is that a set whose pairwise distances are all distinct contains no isosceles triple, linking this question to distinct-distance problems. Erdős Problems #1207
In the plane, the lower-bound approach uses the scarcity of isosceles triples: bound their number in an arbitrary point set, then randomly retain points and delete one point from each surviving isosceles triple. The resulting subset has no such triple. For upper bounds, the challenge is different: construct point sets in which every sufficiently large subset necessarily contains an isosceles triple. Erdős Problems #1207 Isosceles Triangles Determined by a Planar Point Set
References
P. Erdős, A survey of problems in combinatorial number theory, Annals of Discrete Mathematics 6 (1980), 89–115.
Progress summary
Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.
Solutions 0
No solutions have been posted yet.