Multivariate Estimation of Distribution Algorithm with Laplace Transform Archimedean Copula
Ying Gao · 2009
Estimation of distribution algorithm is a new class of evolutionary algorithms. It builds a probability model of promising solutions and samples new individuals from the model. In this paper, we propose a new EDA in which the copula theory is applied to constitute the probabilistic model in the conventional multivariate EDAs. The proposed algorithm employs firstly kernel estimation method to estimate the marginal distributions from the selected parent individuals. Then, the marginal distributions are used to estimate the parameter of the Archimedean copula function generator by using the maximum likelihood method. Finally, according to the multivariate Archimedean copula sample algorithm the new individuals are generated by sampling then dimensional Laplace transform Archimedean copula. The proposed algorithm is applied to some well-known benchmarks. The relative experimental results show that the algorithm has better search ability than original version of estimation of distribution algorithm.