Equilibrium States for Random Non-uniformly Expanding Maps

Alexander Arbieto,Carlos Matheus,Krerley Oliveira
DOI: https://doi.org/10.1088/0951-7715/17/2/013
2003-07-05
Abstract:We show that, for a robust ($C^2$-open) class of random non-uniformly expanding maps, there exists equilibrium states for a large class of <a class="link-external link-http" href="http://potentials.In" rel="external noopener nofollow">this http URL</a> particular, these sytems have measures of maximal entropy. These results also give a partial answer to a question posed by Liu-Zhao. The proof of the main result uses an extension of techniques in recent works by Alves-Araújo, Alves-Bonatti-Viana and Oliveira.
Dynamical Systems
What problem does this paper attempt to address?
The problem that this paper attempts to solve is about the existence of equilibrium states in random non - uniformly expanding maps. Specifically, the authors studied a class of robust ($C^2$ - open) random non - uniformly expanding maps and proved that for a large class of potential functions, these systems have equilibrium states. In particular, these systems have measures of maximal entropy. ### Problem Background In dynamical systems and ergodic theory, equilibrium states are invariant probability measures in the phase space that maximize a certain variational principle (similar to the Gibbs free energy in statistical mechanics). For uniformly hyperbolic systems, the Sinai - Ruelle - Bowen theory has already given existence and uniqueness results of equilibrium states. However, when the system is no longer strictly uniformly hyperbolic, it becomes very challenging to extend this theory. ### Main Contributions of the Paper 1. **Existence of Equilibrium States for Random Non - uniformly Expanding Maps**: - The authors proved that for a $C^2$ - open set $F$ of non - uniformly expanding local diffeomorphisms, if the variation of the potential function $\varphi$ is not large (i.e., low - variation potential), and $f:\Omega\rightarrow F$ is a random map, then the corresponding random system has an equilibrium state. 2. **Existence of Measures of Maximal Entropy**: - As a special case of the above result, the authors proved that these systems have measures of maximal entropy. 3. **Partially Answering the Question Proposed by Liu - Zhao**: - The authors' work partially answers a question proposed by Liu and Zhao, that is, whether there are equilibrium states under more general conditions. ### Main Methods and Techniques - **Technical Extension**: The authors utilized the techniques in the recent works of Alves - Araújo, Alves - Bonatti - Viana and Oliveira et al. - **Entropy Theory of Random Dynamical Systems**: Used the definition of Kolmogorov - Sinai entropy of random dynamical systems and its related theorems. - **Pressure Map and Variational Principle**: Introduced the pressure map $\pi_f$ and used the variational principle to prove the existence of equilibrium states. ### Formula Summary 1. **Definition of Low - variation Potential**: \[ \|\varphi\|_1 < \pi_f(\varphi)-\rho_0h_{\text{top}}(f) \] where $\|\varphi\|_1=\int_\Omega|\varphi_w|_\infty dP(w)$, $\pi_f(\varphi)$ is the pressure map, and $h_{\text{top}}(f)$ is the topological entropy. 2. **Definition of Pressure Map**: \[ \pi_f(\varphi)=\lim_{\varepsilon\rightarrow0}\limsup_{n\rightarrow +\infty}\frac{1}{n}\int_\Omega\log\pi_f(\varphi)(w, n,\varepsilon)dP(w) \] 3. **Variational Principle**: \[ \pi_f(\varphi)=\sup_{\mu\in M(\Omega\times M)}\left\{h_\mu(f)+\int\varphi d\mu\right\} \] Through these formulas and theories, the authors successfully proved the existence of equilibrium states in random non - uniformly expanding maps, thus providing new perspectives and tools for the dynamical study of non - uniformly hyperbolic systems.