The multipole Lempert function is monotone under inclusion of pole sets (Q874372): Difference between revisions

Let \(D\) be a domain in \(\mathbb C^n\) and let \(A\) be a countable subset of \(D\). A function \(\mathbf p:D\to \mathbb R_+\) is called a pole function for \(A\) in \(D\) if \(A=\{a\in D: \mathbf p(a)>0\}\). For a subset \(B\) of \(A\), let \(\mathbf p_B\) be the pole function for \(B\) which agree with the restriction of \(\mathbf p\) to \(B\). The Lempert function with \(\mathbf p\)-weighted poles at \(A\) is defined by \[ \ell_D(\mathbf p,z) = \inf \Biggl\{\prod_{j=1}^l| \lambda_j| ^{\mathbf p(a_j)}\Biggr\}, \] where the infimum is taken over all subsets \(\{\lambda_j\}\) of \(\mathbb D\) (unit disk in the complex plane) for which there is an analytic \(\phi :\mathbb D \to D\) with \(\phi(0)=z\) and \(\phi(\lambda_j)=a_j\) for all \(j\). In the paper under review, the authors apply the Arakelian's approximation theorem in a smart way, taking care of both approximation and interpolation simultaneously, to prove the main result of the paper, which is \[ \ell_D (\mathbf p,\cdot) = \inf \{\ell_D(\mathbf p_B, \cdot ): \emptyset \neq B \;\;\text{a finite subset of }A\}. \] Consequently, the Lempert function is monotone under inclusion of pole sets and \[ \ell_D (\mathbf p,\cdot) = \inf \{\ell_D(\mathbf p_B, \cdot ): \emptyset \neq B \subset A\}. \]