This paper deals with using density ensembles methods to enhance continuous estimation of distribution algorithms. In particular, two density ensembles methods are applied: one is resampling method and the other is subspaces method. In resampling continuous estimation of distribution algorithms, a population of densities are obtained by resampling operator and density estimation operator, and new candidate solutions are reproduced by sampling from all obtained densities. In subspaces continuous estimation of distribution algorithms, a population of densities are obtained by randomly selecting a subset of all variables and estimating the density of high quality solutions in this subspace. The above steps iterate and many densities of high quality solutions in different subspaces are achieved. New candidate solutions are reproduced through perturbing old promising solutions in these subspaces. © 2009 IEEE.
|Title of host publication||Proceedings of the 2009 International Conference on Machine Learning and Cybernetics|
|Publication status||Published - 2009|
- Estimation of distribution algorithms