Growth of bilinear maps

Abstract

For a bilinear map $\ast :{\mathbb{R}}^d\times {\mathbb{R}}^d\to {\mathbb{R}}^d$ of nonnegative coefficients and a vector $s\in {\mathbb{R}}^d$ of positive entries, among an exponentially number of ways combining $n$ instances of $s$ using $n-1$ applications of $\ast$ for a given $n$, we are interested in the largest entry over all the resulting vectors. An asymptotic behavior is that the $n$-th root of this largest entry converges to a growth rate $\lambda$ when $n$ tends to infinity. In this paper, we prove the existence of this limit by a special structure called linear pattern. We also pose a question on the possibility of a relation between the structure and whether $\lambda$ is algebraic.

12 pages, 1 figure; several minor revisions before publication