Random dynamical systems with positive entropy imply second type of distributional chaos (Q2139619)

In this paper the complicated behavior of orbits of random dynamical systems is investigated. The author proves the existence of DC2 chaos using the techniques given in [\textit{W. Huang} and \textit{K. Lu}, Commun. Pure Appl. Math. 70, No. 10, 1987--2036 (2017; Zbl 1377.37110)], where DC2 chaos is one of three variants of so-called distributional chaos, a natural extension of Li-Yorke chaos. Let \(\phi(n,\omega, x)\) be a continuous random dynamical system generated by a continuous random map \(\phi(\omega)(x):= \phi(1,\omega)(x)\) defined as \[ \phi(n,\omega, x)=\left\{ \begin{array}{ll} \phi(\theta^{n-1}\omega)\cdots \phi(\omega)(x), & {n>0;} \\ I, & {n=0.} \\ \end{array} \right. \] The main result of this study is the following theorem (see in the paper for more details about the notation). Theorem. Assume that \(\phi\) is a continuous random dynamical system on a Polish space \(X\) over an ergodic Polish system \((\Omega, \mathcal{F}, P, \theta)\). Let \(K\) be a \(\phi\)-invariant random set with compact \(\omega\)-section \(K(\omega)\) such that \(\phi(\omega)(K(\omega)) = K(\theta\omega)\). If \(h_{\mathrm{top}}(\phi, K)> 0\), then there exists a subset \(\Omega_1\) of \(\Omega\) with \(P(\Omega_1) = 1\) such that for any \(\omega\in \Omega_1\), there is a strong DC2 chaotic set \(S(\omega)\subset K(\omega)\) of a union of countably many Cantor sets for \((\omega,\phi)\). Some known results on the relative Pinsker \(\sigma\)-algebra for a factor map between two invertible Lebesgue systems are given.