A Vitali set can be homeomorphic to its complement (Q2460645): Difference between revisions

The main result of the paper says that there exists a~homeomorphism \(F:2^\omega\to2^\omega\) and a~set \(V\subseteq2^\omega\) such that \(F[V]=2^\omega\setminus V\) and \(V\)~is both a~Vitali set and a~Bernstein set. Moreover, every non-cancelable composition of a finite number of copies of~\(F\) and translations of~\(2^\omega\) has no fixed points. On the other hand there is a~Bernstein Vitali set \(B\subseteq\mathbb R\) such that \(F[B]=\mathbb R\setminus B\) for no Borel isomorphism~\(F:\mathbb R\to\mathbb R\). The proofs use the technique of paradoxical decompositions and the idea of the Banach-Tarski paradox. These results can be considered as an approach to the open problem of the existence of a~Hamel base \(H\subseteq\mathbb R\) and a~homeomorphism \(f:\mathbb R\to\mathbb R\) such that \(f[H]=\mathbb R\setminus H\).