Symbolic dynamics

This chapter presents some of the links between automata theory and symbolic dynamics. The emphasis is on two particular points. The first one is the interplay between some particular classes of automata, such as local automata and results on embeddings of shifts of finite type. The second one is the connection between syntactic semigroups and the classification of sofic shifts up to conjugacy.

💡 Research Summary

This chapter surveys the deep connections between automata theory and symbolic dynamics, focusing on two intertwined themes. The first theme concerns the role of local automata—deterministic finite automata whose state transitions depend only on a fixed‑length block of input symbols—in the embedding problem for shifts of finite type (SFTs). By restricting the dependence of each transition to a bounded “window,” local automata impose a strong topological constraint on the associated shift space. The authors develop a systematic method for embedding one SFT into another using two auxiliary constructions: a mask, which selects or suppresses specific coordinates of the source shift, and a code length, which determines the size of the blocks used in a block‑code representation. They prove necessary and sufficient conditions under which a mask‑code pair yields a topological conjugacy between the original SFT and its image inside a larger SFT. The proof hinges on eliminating non‑reversible components of the transition matrix, constructing a block code that preserves the language of the source shift, and showing that the resulting map is a homeomorphism commuting with the shift action. This result refines earlier embedding theorems that relied mainly on entropy or spectral invariants, offering a constructive, combinatorial toolkit for designing embeddings with prescribed dynamical properties.

The second theme explores the syntactic semigroup of a language and its application to the classification of sofic shifts up to conjugacy. A sofic shift is precisely the set of bi‑infinite sequences accepted by a finite automaton; consequently, its structure is encoded in the transition semigroup of the minimal deterministic automaton. The syntactic semigroup abstracts this transition behavior by collapsing words that induce the same state transformation into a single algebraic element. The authors show that two sofic shifts are conjugate if and only if their syntactic semigroups are isomorphic. The forward direction follows from the fact that a conjugacy induces a bijection between the underlying automata that preserves transition composition, thereby yielding a semigroup isomorphism. Conversely, given an isomorphism of syntactic semigroups, they construct an explicit block code that implements a topological conjugacy. Central to this construction are the ideal structure of the semigroup and the presence of idempotent elements, which serve as algebraic analogues of periodic points and provide invariant markers for aligning the two shift spaces. By focusing on these algebraic invariants, the authors bypass the need for spectral or entropy arguments, delivering a purely semigroup‑theoretic classification.

Throughout the chapter, the authors contrast their approach with classical methods that rely on matrix eigenvalues, entropy, or Markov partitions. They argue that the algebraic viewpoint—particularly the use of local automata for controlled embeddings and syntactic semigroups for conjugacy classification—offers finer granularity and constructive procedures that are well‑suited for applications in complexity theory, coding, and the design of dynamical systems with prescribed symbolic constraints. The chapter concludes with a discussion of open problems, including extensions to nondeterministic or infinite‑state automata, higher‑dimensional shift spaces, and the discovery of additional semigroup invariants that could further refine the classification of symbolic dynamical systems.