Rate distortion dimension of random Brody curves

Abstract

The main purpose of this paper is to propose an ergodic theoretic approach to the study of entire holomorphic curves. Brody curves are one-Lipschitz holomorphic maps from the complex plane to the complex projective space. They naturally form a dynamical system, and "random Brody curves" in the title refers to invariant probability measures on it. We study their geometric and dynamical properties. Given an invariant probability measure $μ$ on the space of Brody curves, our first main theorem claims that its rate distortion dimension is bounded by the integral of a "potential function" over $μ$. This result is analogous to the Ruelle inequality of smooth ergodic theory. Our second main theorem claims that there exists a rich variety of invariant probability measures attaining equality in this "Ruelle inequality for Brody curves". The main tools of the proofs are the deformation theory of Brody curves and the variational principle for mean dimension with potential. This approach is motivated by the theory of thermodynamic formalism for Axiom A diffeomorphisms.

Figures (2)

• Figure 1: Here $P$ has four vertexes (denoted by dots), four $1$-dimensional simplexes and one $2$-dimensional simplex. The points $b$ and $d$ are vertexes of $P$ whereas $a$ and $c$ are not. We have $\dim_a P = \dim_b P =2$ and $\dim_c P = \dim_d P =1$.
• Figure 2: The big square is $[0, L_n)^k$ and small squares are $v+\lambda + [0, M)^k$ ($\lambda \in \Lambda_n$). The shadowed region is $E_{n,v}$.

Theorems & Definitions (69)

• Example 2.1
• Example 2.3
• Theorem 2.4
• Theorem 2.5
• Remark 2.6
• Remark 2.7
• Example 2.8: Continuation of Example \ref{['example: random Brody curves']}
• Theorem 4.1
• Lemma 4.2
• Lemma 4.3: Subadditivity of mutual information