Multiple ergodic averages for variable polynomials (Q2158228)

The author establishes limit theorems for multiple ergodic averages for variable polynomials, subject to some technical conditions. To give an example, Theorem 1.2 states that for a strongly independent \(\ell\)-tuple of variable polynomials \((p_{1,N},\dots, p_{\ell,N})\) (see the paper for a definition), and for any ergodic system \((X,\mathcal{B},\mu,T)\) and bounded functions \(f_1,\dots, f_\ell\) one has \[ \lim_{N \to \infty} \frac{1}{N} \sum_{n=1}^N T^{[p_{1,N}(n)]} f_1 \cdot \ldots \cdot T^{[p_{\ell,N}(n)]} f_\ell = \int f_1 d\mu \cdot \ldots \cdot \int f_\ell d\mu. \] The results in the paper address a question of \textit{N. Frantzikinakis} [Bull. Hell. Math. Soc. 60, 41--90 (2016; Zbl 1425.37004)], and stand in the tradition of Furstenberg's and Bergelson's work on multiple ergodic averages along polynomials.