Dice Question Streamline Icon: https://streamlinehq.com

Equivalence of complete and partial automaton semigroups/monoids

Determine whether every (partial) automaton semigroup or automaton monoid—generated by a non-complete deterministic letter-to-letter transducer—can also be generated by some complete deterministic letter-to-letter transducer; equivalently, ascertain whether the class of complete automaton semigroups and monoids coincides with the class of partial automaton semigroups and monoids.

Information Square Streamline Icon: https://streamlinehq.com

Background

The paper works with complete deterministic letter-to-letter transducers and their generated automaton semigroups/monoids, while noting that these notions extend to non-complete automata (yielding partial automaton semigroups/monoids). Whether every partial instance can be represented by a complete one is a foundational unresolved question about the scope of the model.

Resolving this would clarify the relationship between two widely used definitions in the literature and determine if results stated for complete automata automatically cover the partial setting.

References

It is not known whether the two classes coincide (and we refer the reader to [structurePart] for more details on this question and the general concepts).

The Freeness Problem for Automaton Semigroups (2402.01372 - D'Angeli et al., 2 Feb 2024) in Preliminaries, Automaton Semigroups and Monoids, Remark (first)