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.
494 of 1,221 problems match
Every problem posed by Paul Erdős, with its status, references, discussion and proof claims.
Asks whether bounded clique number and large chromatic number force two anticomplete vertex sets, both of large chromatic number. This is the El-Zahar–Erdős problem, open beyond the case of chromatic number .
Asks whether some Sierpinski number, an odd with never prime, has no finite set of primes containing a divisor of every .
Asks for the shortest length of a path from to inside , where is a monic polynomial of degree with all roots in .
Asks for the minimum over of , where the are the Lagrange basis polynomials, and whether .
For a sequence of nodes in , let . Asks whether some has for infinitely many , and whether for almost all .
Asks whether every orbit of the shortcut Collatz map reaches ; open, with verification below , no nontrivial cycle with at most local minima, and Tao's almost-all theorem. erdosproblems.com adds a caveat on the reported Erdős prize figure.
Asks whether infinitely many , or any , have prime for every with .
Asks to estimate, for given primes , how many integers divisible by some every interval of consecutive positive integers must contain.
Asks whether whenever and are sets of positive integers with and contains all large integers.
Asks whether is an essential component, that is, a set with whenever the Schnirelmann density lies strictly between and .
Asks whether, for every , some ring or field in has Hausdorff dimension . Mauldin's theorem gives yes under the continuum hypothesis, so ZFC cannot refute the existence.
Asks whether the chromatic number of a random graph on vertices, each edge present independently with probability , is almost surely concentrated on boundedly many values, and whether it is within a slowly growing of any given with probability below .
Asks for the most edges an -graph on vertices can have with no vertices spanning edges, the Brown–Erdős–Sós problem. The associated conjecture is proved for and, in its linear form, for large uniformity, but open for -graphs with and . A 2026 preprint claims the asymptotics whenever is an integer with , except for -graphs with even .
Asks whether Erdős's 1964 upper exponent for the Turán number of , the complete -partite -uniform hypergraph with vertices per class, is attained up to . It is known for and , and a 2026 preprint claims the case , .
Asks whether there is a constant such that every finite projective plane has a set of points meeting each line in at least and at most points.
Asks for an asymptotic formula for the number of subgroups of the symmetric group , and for statistical results on their orders.
Asks for a statistical description of the arithmetic structure of the orders of subgroups of the symmetric group .
Asks whether, for infinite , implies ; open under the Erdős–Hajnal list's conditions and all , which exclude the failures of the erdosproblems.com wording.
Asks for a proof of without the generalized continuum hypothesis.
Asks whether several partition relations for pairs on the ordinals and hold under the generalized continuum hypothesis, or are consistent with it.