Documentation

FormalConjectures.Wikipedia.CernyConjecture

Černý Conjecture #

A synchronizing word (also called a reset word) for a deterministic finite automaton (DFA) $M = (Q, \Sigma, \delta)$ is a word $w \in \Sigma^*$ such that reading $w$ from any state always leads to the same single state — formally, $\exists p \in Q, \forall q \in Q, \delta^*(q, w) = p$.

A DFA is called synchronizing if it admits at least one synchronizing word.

The Černý conjecture asserts that every synchronizing DFA with $n$ states has a synchronizing word of length at most $(n - 1)^2$. This bound is sharp: the family of Černý automata $C_n$ witnesses it, requiring exactly $(n - 1)^2$ steps.

Status: Open. The best known upper bound is $\left(\frac{7}{48} + \frac{2 \cdot 15625}{1597536}\right) n^3 + o(n^3) \approx 0.1654\,n^3$ (Shitov, 2019). The bound $(n - 1)^2$ has been verified for small $n$ and for special classes of automata (e.g., Eulerian, aperiodic, cyclic automata).

We use Mathlib's DFA α σ (from Mathlib.Computability.DFA), together with the auxiliary DFA.IsSynchronizingWord and DFA.IsSynchronizing predicates defined in FormalConjecturesForMathlib.Computability.DFA.

References:

theorem CernyConjecture.cerny_conjecture :
True ∀ {α : Type u_1} {σ : Type u_2} [inst : Fintype σ] (M : DFA α σ), M.IsSynchronizing∃ (w : List α), M.IsSynchronizingWord w w.length (Fintype.card σ - 1) ^ 2

Černý Conjecture: Every synchronizing DFA with $n$ states admits a synchronizing word of length at most $(n - 1)^2$.

theorem CernyConjecture.shitov_upper_bound :
∃ (f : ), (f =o[Filter.atTop] fun (n : ) => n ^ 3) ∀ {α : Type u_1} {σ : Type u_2} [inst : Fintype σ] (M : DFA α σ), M.IsSynchronizing∃ (w : List α), M.IsSynchronizingWord w w.length (7 / 48 + 2 * 15625 / 1597536) * (Fintype.card σ) ^ 3 + f (Fintype.card σ)

Shitov's bound (2019): Every synchronizing DFA with $n$ states admits a synchronizing word of length at most $\left(\frac{7}{48} + \frac{2 \cdot 15625}{1597536}\right) n^3 + o(n^3)$, where the $o(n^3)$ term is uniform over all alphabets. This is the best known upper bound towards the Černý conjecture.