Conjectures about Latin Squares
Conjecture 3.2 in [Wa2011]: Each Latin square of odd order has at least one transversal.
Each problem states how progress is verified and what counts as a contribution. Besides the problems curated here, the catalogue includes open conjectures from Formal Conjectures (with Lean statements), optimization constants and the AlphaEvolve problems. Know one that belongs here? Propose a problem.
757 shown· page 3 of 16
Conjecture 3.2 in [Wa2011]: Each Latin square of odd order has at least one transversal.
For any odd natural number p if two of the following conditions hold, then all three must hold: 1. 2^p-1 is prime 2. (2^p+1)/3 is prime 3. Exists a number k such that p = 2^k pm 1 or p = 4^k pm 3
The MLC conjecture, stating that the mandelbrot set is locally connected.
Problem 2 in [Ar2013]: Give an example in ZFC of a weakly first- countable compact Hausdorff space X such that 𝔠 < |X|. Note: [Ar2013] uses a blanket convention that all spaces are Tychonoff and "compact" means compact Hausdorff.
Problem 13 in [Ar2013]: Is it true that every infinite homogeneous compact hausdorff space contains a non-trivial convergent sequence?
All terms of A038552 are congruent to 19 pmod24.
All members of the sequence satisfy n ≡ 108 pmod216.
For members of the sequence other than 8, we have k + 1 is prime.
After a(2) = 5, is there another prime?
A100800 Conjecture: No term is zero.
It is conjectured k always exists.
Cormier and Selfridge found 5 starting values for which the sequences appear to not merge. The sequences were checked up to 10^8.
Conjecture: a(2) and a(121) are primes. Are there any more?
Does the sequence contain every positive integer (cf. A169741)?
Conjecture: There are infinitely many primes in this sequence.
a(n) = 0 for n = 1, 6, 30 and 54. Are there any others?
Conjecture: a(n) > 0 for n > 3.
Conjecture: a(n) > 0 for n > 3.
Do the absolute values cover A004275? A004275 is 1 together with the nonnegative even numbers. The conjecture asks whether every member of A004275 occurs as |a(n)| for some term of the sequence.
The Flint Hills series summing csc(n)^2 / n^3 from n=1 to ∞ converges. (Note that we 0-index the series below.)
Does there exist an undirected graph with 99 vertices, in which each two adjacent vertices have exactly one common neighbor, and in which each two non-adjacent vertices have exactly two common neighbors?
This sequence is believed to be infinite.
Jones's conjecture (first kind): for every positive integer k, there are infinitely many primes p that start a first-kind Cunningham chain of exactly length k.
De Giorgi's conjecture holds in dimension n ≤ 8.
Conjecture 1.1 (Dean, 1988). For every integer k ≥ 3, every finite simple graph with minimum degree at least k contains a cycle whose length is divisible by k. A cycle has length at least 3, so the divisor is never 0 and the statement is not satisfied for a trivial reason.
"a(31) = a(177147) = 311. Is there any solution to a(n) = n? - _Franklin T. Adams-Watters_, Dec 18 2006"
No closed-form expression that allows efficient computation of Dedekind numbers is currently known.
If a(n) is in A005153, then n is in A005153. - Jaycob Coleman, Sep 27 2014 We require 0 < n because a(0) = 1 is in A005153 (practical numbers), but 0 is not.
Conjecture: a(n) = primorial(n) for infinitely many n.
Conjecture I: if n > 2, then a(A005382(n))/12 is prime, where A005382 is the sequence of primes p such that 2p-1 is also prime. Since A005382(1) = 2, A005382(2) = 3 and A005382(3) = 7, this says that a(p)/12 is prime for every prime p > 3 such that 2p-1 is also prime.
"I conjecture that a(4) is the only zero. - _Jon Perry_, Mar 22 2004" Stated as a biconditional: the claim that a(4) is the only zero asserts both that a(4) = 0 and that no other index vanishes. A bare implication a n = 0 → n = 4 would be satisfied vacuously by a sequence with no zero at all.
Conjecture: a(n) = 0 for no n > 28. - _Zhi-Wei Sun_, Aug 26 2013
Does the determinant of the sum A + B of two n × n normal complex matrices A and B always lie in the convex hull of the n! points Π_i (λ(A)_i + λ(B)_σ(i))? Here the numbers λ(A)_i and λ(B)_i are the eigenvalues of A and B, and σ is an element of the symmetric group S_n.
This suggests the ratio is approaching a limit close to 0.87. Formalized as: The sequence of ratios P(N)/Neg(N) converges to a limit L, and L is in the interval (0.8, 0.9).
Dickson's conjecture If a finite set of linear integer forms f_i(n) = a_i n+b_i satisfies Schinzel condition, there exist infinitely many natural numbers m such that f_i(m) are primes for all i.
In this powers of 2 sequence, does 1 occur infinitely often?
"This sequence is positive on average, since 1/log(3) > 1/log(4). Do all integers appear infinitely often?" - Charles R Greathouse IV, Feb 07 2013
For n > 8, 2^n is not the the sum of distinct powers of 3. Expressed here in terms of the base 3 digits of n. This conjecture is equivalent to the halting of a 15-state 2-symbol Turing Machine. TODO(lezeau): Formalize the Turing Machine version of this problem.
a(0), a(1), a(5), a(6), a(7) and a(11) are primes. Are there any more?
The "strong Diophantine 5-tuple conjecture", so-called because it implies the Diophantine 5-tuple theorem (see noIntegralDiophantineFiveTuple_of_hasUniqueExtensionOfForall). [Du]
All members of the sequence A56777 come from prime quadruples.
Every even number greater than 4208 is the sum of two twin primes.
The Elliott–Halberstam conjecture: for every θ < 1 and A > 0 there exists a constant C > 0 such that Σ_1 ≤ q ≤ x^θ E(x; q) ≤ C x/log^A x for all x > 2.
Equational Theories, Problem 8.1. Does Equation 677 imply Equation 255 in every finite magma? The project tentatively conjectures that the answer is no; a false answer is equivalent to the existence of a finite countermodel satisfying Equation 677 but not Equation 255.
The sequence (3/2)^n is equidistributed modulo 1.
Is there some k such that every large integer is the sum of a prime and at most k powers of 2?
Is the diameter of A at least Cn for some constant C > 0?
For any 0<α<1, let f(α,n)=1/log nΣ_1≤ k≤ n(1/2- α k). Does f(α,n) have an asymptotic distribution function? In other words, is there a non-decreasing function g such that g(-∞)=0, g(∞)=1, and lim_n→ ∞lvert α∈ (0,1): f(α,n)≤ crvert=g(c)?
Are there infinitely many solutions to φ(n) = φ(n+1), where φ is the Euler totient function?
For any fixed c > 0, if x is sufficiently large then there exists n ≤ x such that the values of φ(n+k) are all distinct for 1 ≤ k ≤ (log x)^c. This is an open problem.