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.
110 of 1,221 problems match
Every problem posed by Paul Erdős, with its status, references, discussion and proof claims.
Asks whether the number of -element sets needed to force a -sunflower is at most exponential in , that is, below for a constant depending on .
Asks whether the edges of every chordal graph on vertices can be partitioned into cliques.
Asks whether every convex polygon has a vertex with no four other vertices equidistant from it. Erdős first asked it with three in place of four, and Danzer's convex nonagon refutes that version.
Asks for an estimate of , the least number of Abelian subgroups that suffices to cover every group in which any set of more than elements contains two distinct commuting elements.
Asks for a constant with , where is the least such that every -coloring of has vertices with no triangle in some color. A random coloring refutes the erdosproblems.com wording, and no source gives another intended form.
Asks whether a countably infinite set with for all distinct and all integers must be sparse.
Asks whether every finite coloring of has arbitrarily large finite sets whose sums and products of distinct elements all have one color.
Asks for good upper bounds on , the least such that every has a -term arithmetic progression with , in particular whether every has some with .
Asks whether for the -dimensional hypercube graph , which has vertices. It was open on erdosproblems.com, and a 2026 OpenAI release preprint claims the linear bound in full.
Asks whether must be irrational whenever , where are the Fibonacci numbers.
Asks for the maximum density of integers covered by choosing one congruence class for each modulus in a finite set, and whether equal classes minimize the density.
Asks whether for every one can choose a class for each prime so that every large integer is for some prime and integer .
Asks whether, for every large , one can write as the sum of over distinct integer intervals, none overlapping or adjacent and each of length at least .
Asks for estimates of , the size of the largest in which no is a sum of reciprocals of other distinct members, and whether .
Asks for estimates of , the size of the largest with no distinct satisfying , and whether .
Asks whether every positive rational with squarefree denominator is a sum of distinct unit fractions whose denominators are each a product of two distinct primes.
Asks whether some minimal basis of order , with elements , has tending to a nonzero constant; a 2026 proof claim is pending. Erdős first asked this for any basis of order , which Cassels answered yes.
Asks how large can be if for all distinct , or if instead .
Asks for , where is the largest finite exact order of an additive basis of order .
Asks for which a complete sequence can stay complete after removing any elements yet fail to be complete after removing any elements.
Asks whether there is a constant , and what it is, with , where is the least such that with .