Černý’s conjecture

For every synchronizing deterministic finite automaton A=(Q,Σ,δ)\mathcal{A}=(Q,\Sigma,\delta) with ∣Q∣=n|Q|=n, there exists a word w∈Σ∗w\in\Sigma^* and a state q0∈Qq_0\in Q such that δ(q,w)=q0\delta(q,w)=q_0 for every q∈Qq\in Q and ∣w∣≤(n−1)2|w|\leq (n-1)^2.

References

Primary source

arXiv

Additional references

Progress summary

Refreshed
Open

The conjecture remains unproved in general: only special types of automata are known to obey the proposed quadratic bound, and no counterexample is known.

Posed by Černý in 1964, the conjecture says that every synchronizing automaton with nn states has a reset word of length at most (n−1)2(n-1)^2. The matching lower bound is attained by Černý’s automata, but the general upper bound remains unknown.

Known results

  • Aperiodic automata: bound n(n−1)/2n(n-1)/2 (2007); Volkov improved this to n(n+1)/6n(n+1)/6 for strongly connected cases.
  • Eulerian synchronizing automata: bound n2−3n+3n^2-3n+3.
  • 11-extensible and one-cluster synchronizing automata satisfy the conjecture.
  • The best general upper bound is cubic, due to Shitov: C(n)≤(748+15625798768)n3+o(n3)\mathfrak{C}(n)\leq\left(\frac{7}{48}+\frac{15625}{798768}\right)n^3+o(n^3); no quadratic general bound is known.

August 2026 survey

A new survey consolidates open problems and auxiliary results but reports no proof or refutation. Earlier claimed proofs by Trahtman were undermined by his February 10, 2024 acknowledgment of an error; the latest survey continues to classify the conjecture as open.

Current status (as of August 2026): Černý’s conjecture remains open; the quadratic bound is proved for several restricted classes, while the general problem has only a cubic upper bound and no known counterexample.

Sources

Solutions 0

No solutions have been posted yet.