Geometric properties of domains related to \(\mu\)-synthesis (Q2348405)

The author studies the holomorphic geometry of the \textit{generalized tetrablock} \[ \mathbb E_E=\mathbb E_{n;s;r_1 ,\dots,r_s}:=\{x\in\mathbb C^N: \forall_{z\in\overline{\mathbb D}^s}: R_x(z)\neq0\}, \] where \[ R_x(z):=1+\sum_{j=1}^N(-1)^{|\alpha^j|}x_jz^{\alpha^j},\quad x=(x_1,\dots,x_N)\in\mathbb C^N,\; z=(z_1,\dots,z_s)\in\mathbb C^s, \] \(n, s, r_1,\dots,r_s\in\mathbb N\), \(n\geq2\), \(r_1+\dots+r_s=n\), \(\{\alpha^1,\dots,\alpha^N\}:=\mathbb N_0^s\cap([0,r_1]\times\dots\times[0,r_s])\setminus\{(0,\dots,0)\}\), \(N:=(r_1+1)\cdots(r_s+1)-1\), \(\alpha^1<\dots<\alpha^N\) with the order \(<\) defined by the relation \(\alpha<\beta: \Longleftrightarrow \alpha_{j_0}<\beta_{j_0}\) with \(j_0:=\max\{j: \alpha_j\neq\beta_j\}\). The set \(\mathbb E_E\) is a simultaneous generalization of many classical domains. For example: \(\mathbb E_{2;1;2}=\mathbb G_2=\) the symmetrized bidisc, \(\mathbb E_{n;1;n}=\mathbb G_n=\) the symmetrized polydisc, \(\mathbb E_{2;2;1,1}=\mathbb E=\) the tetrablock, and \(\mathbb E_{n;2;n -1,1}=\mathbb E_n=\) the \(\mu_{1,n}\)-quotient. In particular, the author proves the following results: {\parindent=6mm \begin{itemize} \item[-] \(\mathbb E_E\) is a bounded \((|\alpha^1|, \dots, |\alpha^N|)\)-balanced domain. \item [-] If there exists a \(j\) such that \(r_j\geq3\), then the Carathéodory and Lempert functions do not coincide; consequently, from the point of view of the Lempert theorem the only interesting examples are \(\mathbb E_{n;s;r_1,\dots,r_s}\) with \(r_1,\dots,r_s\in\{1,2\}\). \item [-] \(\mathbb E_E\) cannot be exhausted by domains biholomorphic to convex ones. \item [-] \(\mathbb E_E\) is not circled. \item [-] If there is a \(j\) such that \(r_j\geq3\), then \(\mathbb E_E\) is neither \(\mathbb C\)-convex nor star-like at the origin. \item [-] If \(r_2=\dots=r_s=1\), then \(\mathbb E_E\) is linearly and polynomially convex. \end{itemize}}