Resumo

SALINAS GUTIERREZ, Rogelio; HERNANDEZ AGUIRRE, Arturo  e  VILLA DIHARCE, Enrique Raúl. A extension for MIMIC algorithm through Copulas. Nova scientia [online]. 2010, vol.2, n.3, pp.01-13. ISSN 2007-0705.

A new way of modeling probabilistic dependencies in Mutual Information Maximization for Input Clustering (MIMIC) algorithm is presented. By means of copulas it is possible to separate the dependence structure from marginal distributions in a joint distribution. The use of copulas as a mechanism for modeling distributions and its applicaction to MIMIC is illustrated on Rosenbrock test function.

Palavras-chave : Global optimization; Evolutionary computing; Estimation of Distribution Algorithms; Copulas.