neo2018

Speakers - Manuel Bogoya

A generalization of the averaged Hausdorff distance $\Delta_p$
Manuel Bogoya
Pontificia Universidad Javeriana, Colombia

Talk Abstract: The averaged Hausdorff distance $\Delta_p$ is an inframetric which has been recently used in evolutionary multiobjective optimization (EMO). We introduce a new two-parameter performance indicator $\Delta_{p,q}$ which generalizes $\Delta_p$ as well as the standard Hausdorff distance. For $p,q \geq 1$ the indicator $\Delta_{p,q}$ (that we call the $(p,q)$-averaged distance) turns out to be a proper metric and preserves some of the $\Delta_p$ advantages. We show several properties of $\Delta_{p,q}$, including a geometrical role for $p$ and $q$, and provide a comparison with $\Delta_p$ and the standard Hausdorff distance.

Bio: Manuel Bogoya received his PhD in operator theory from CINVESTAV research center in Mexico (2010). His research also includes statistical models for education, spectral distribution, and Toeplitz operators. He is currently associate professor at Pontificia Universidad Javeriana, Bogotá-Colombia.