Given an integer nonsingular $n\times bn$ matrix $M$ and a point $y \in
\mathbb{R}^n/\mathbb{Z}^n$, consider the set $\tilde E(M,y)$ of vectors $x\in
\mathbb{R}^n$ such that $y$ is not a limit point of the sequence $\{M^k x \mod
\Z^n: k\in\N\}$. S.G. Dani showed in 1988 that whenever $M$ is semisimple and
$y \in \mathbb{Q}^n/\mathbb{Z}^n$, the set $\tilde E(M,y)$ is winning in the
sense of W.
Given b > 1 and y \in \mathbb{R}/\mathbb{Z}$, we consider the set of $x\in
\mathbb{R}$ such that $y$ is not a limit point of the sequence $\{b^n x \bmod
1: n\in\N\}$. Such sets are known to have full Hausdorff dimension, and in many
cases have been shown to have a stronger property of being winning in the sense
of Schmidt. In this paper, by utilizing Schmidt games, we prove that these sets
and their bi-Lipschitz images must intersect with `sufficiently regular'
fractals $K\subset \mathbb{R}$ (that is, supporting measures $\mu$ satisfying
certain decay conditions).
Given b > 1 and y \in \mathbb{R}/\mathbb{Z}$, we consider the set of $x\in
\mathbb{R}$ such that $y$ is not a limit point of the sequence $\{b^n x \bmod
1: n\in\N\}$. Such sets are known to have full Hausdorff dimension, and in many
cases have been shown to have a stronger property of being winning in the sense
of Schmidt. In this paper, by utilizing Schmidt games, we prove that these sets
and their bi-Lipschitz images must intersect with `sufficiently regular'
fractals $K\subset \mathbb{R}$ (that is, supporting measures $\mu$ satisfying
certain decay conditions).
We exploit dynamical properties of diagonal actions to derive results in
Diophantine approximations. In particular, we prove that the continued fraction
expansion of almost any point on the middle third Cantor set (with respect to
the natural measure) contains all finite patterns (hence is well approximable).
Similarly, we show that for a variety of fractals in [0,1]^2, possessing some
symmetry, almost any point is not Dirichlet improvable (hence is well
approximable) and has property C (after Cassels). We then settle by similar
methods a conjecture of M.