Schmidt's game on fractals (Q836104)

The paper under review considers tuples of simultaneously badly approximable points in \(\mathbb R^N\), \textit{i.e.}, vectors \(x \in {\mathbb R}^n\) such that there is a \(\delta > 0\) with \(d(x,{{1} \over {q}}\mathbb Z^n) \geq \delta q^{-(n+1)/n}\) for any \(n \in \mathbb Z\setminus\{0\}\), where \(d(x,E)\) denotes the Euclidean distance from the point \(x\) to the set \(E\). It was shown by \textit{D. Kleinbock} and \textit{B. Weiss} [Isr. J. Math. 149, 137--170 (2005; Zbl 1092.28004)] and independently by \textit{S. Kristensen}, \textit{R. Thorn} and \textit{S. Velani} [Adv. Math. 203, No. 1, 132--169 (2006; Zbl 1098.11039)] that the intersection of the set of badly approximable vectors with a suitably nice fractal set has maximal Hausdorff dimension. In the present paper, the author extends this by proving that the intersection between the support of an absolutely friendly measure and the set of badly approximable vectors is winning for the so-called Schmidt game. Supports of absolutely friendly measures include attractors of families of contracting similarities satisfying the open set condition, and hence sets such as the Cantor middle third set, the von Koch curve and the Sierpiński gasket. In the case of such attractors, the author deduces that the intersection in question has maximal Hausdorff dimension.