Problems
Every problem posed by Paul Erdős, with its status, references, discussion and proof claims.
Every problem posed by Paul Erdős, with its status, references, discussion and proof claims.
618 of 1,221 problems match
Every problem posed by Paul Erdős, with its status, references, discussion and proof claims.
Asks whether a finite set of integers with must satisfy for some constant .
Asks whether, for every real , there are pairs with and , where is the sum of divisors.
Asks whether there is a constant such that every integer with is a sum of distinct proper divisors of .
Asks whether, for and large , every -uniform hypergraph with chromatic number has at least edges, with a defectively worded equality clause. Alon's counterexamples disprove it, but the case is open.
Asks whether there is an absolute constant such that, for all , every -uniform hypergraph with chromatic number has a vertex in at least edges. Erdős and Lovász proved this, and the page records an explicit constant that works for every .
Asks whether there is a -critical -uniform hypergraph in which every vertex has degree at least .
Asks for estimates of the least such that contains a subset whose product with is a square. Erdős's original question, whether the with have density zero, has the answer yes.
Asks whether disjoint triangles joined by a Hamiltonian cycle on their vertices always give a -colorable graph; proved by Fleischner and Stiebitz.
Asks whether the squares are Ramsey -complete, meaning that in every -coloring of the squares each large enough integer is a sum of distinct squares of one color.
Asks whether the even numbers together with the odd non-squarefree numbers form a largest in which no product with is squarefree.
Asks whether, for each , the integers with and each of the form have density .
Asks whether an infinite plane set in which, for some fixed , every of its points include at least with no three collinear must be a finite union of sets with no three collinear.
Asks whether an infinite set of naturals must be a finite union of sets with no three-term arithmetic progression if, for a fixed , any of its members include at least forming no such progression.
Asks whether, for each , some makes the integers , with having at most prime divisors, have density at least .
Asks for an estimate of the maximum of over with no solution to in which and every prime factor of exceeds .
Asks whether the density of the integers for which is a sum of distinct divisors of satisfies for some constants . A Lean disproof, certified by the bounty site Conjectures.io in 2026, shows that it does not.
Asks how the number of Sidon subsets of compares with , where is the size of the largest Sidon subset.
Asks whether the number of maximal Sidon subsets of is below , and whether it exceeds for some .
Asks whether, for , the largest sets in with at most representations of each sum, and of each positive difference, have sizes and with , and whether .
Asks whether every subset of of size at least , for some constant , contains three distinct members whose pairwise sums also lie in the set. It was proved in a 2026 preprint, developed with GPT-5.5 Pro, which erdosproblems.com accepted.
Asks whether a subset of in which no sum of consecutive members lies in the set has size at most ; false, by Freud's 1993 construction of density .
Asks whether an additive basis of order whose representation counts tend to infinity must contain a minimal such basis.
Asks whether the union of two disjoint additive bases of order must contain a minimal additive basis of order .
Asks whether an additive basis of order with can be split into two disjoint additive bases of order .