On the approximation order of extremal point methods for hyperbolic minimal energy problems

From MaRDI portal
(Redirected from Publication:1769084)





Let \(\Gamma\) be an analytic Jordan curve in the unit disk and let \(\Omega\) be the doubly connected domain bounded by \(\Gamma\) and the unit circle. The hyperbolic energy of \(\Gamma\) is \[ I^h(\Gamma)=-\min_{\mu} \int_\Gamma\int_\Gamma \log \left | \frac{z-\zeta}{1-z\bar{\zeta}}\right | \;d\mu(z)d\mu(\zeta), \] where the minimum is taken over all probability measures on \(\Gamma\). \textit{K. Menke} [see Numer. Math. 54, No. 2, 125--143 (1988; Zbl 0655.30006) and references therein] introduced a special discretization of hyperbolic energy: Consider points (Menke points) \(z_1,\zeta_1, z_2,\zeta_2,\dots, z_n,\zeta_n\) lying on \(\Gamma\) in this order and minimizing the quantity \[ \prod_{j=1}^n\prod_{k=1}^n \left | \frac{z_j-\zeta_k}{1-z_j\bar{\zeta_k}}\right | . \] Menke presented numerical examples that show that one can approximate the hyperbolic energy by using these points. In the paper under review the author makes this approximation precise and rigorous. Moreover, he shows that the Menke points approach the images of roots of unity under the conformal map from the annulus \(\{1<| z| <e^{I^h}\}\) onto \(\Omega\), geometrically fast as \(n\to\infty\). This establishes a conjecture of Menke. Explicit bounds for the error are given and a method for the numerical determination of the minimizing measure is presented. The basic idea in the proofs is to reformulate the minimization problem as a variational problem for univalent functions and use estimates for the coefficients of univalent functions.











This page was built for publication: On the approximation order of extremal point methods for hyperbolic minimal energy problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1769084)