Multivariate growth and cogrowth
Abstract
We investigate a multivariate growth series L( z), z ∈ Cd associated with a regular language L over an alphabet of cardinality d. Our focus is on languages coming from subgroups of the free group and from subshifts of finite type. We develop a mechanism for computing the rate of growth L( r) of L in the direction r ∈ Rd. Using the concave growth condition (CG) introduced by the second author in quint2002divergence and the results of Convex Analysis we represent L( r) = (L( r)) as a support function of a convex set that is a closure of the Relog image of the domain of absolute convergence of L( z). This allows us to compute L( r) in some important cases, like a Fibonacci language or a language of freely reduced words representing elements of a free group F2. Also we show that the methods of the Large deviation theory can be used as an alternative approach. Finally, we suggest some open problems directed on the possibility of extensions of the results of the first author from grigorchuk1980symmetrical on multivariate cogrowth.
Turn this paper into a lesson
ArcXiv compiles a structured reading guide from this paper's metadata: plain-English importance, contributions, prerequisite concepts, which sections to read first, flashcards, and a quiz. Grounded in the abstract, never invented.