Surrogate modeling approximation using a mixture of experts based on Expectation-Maximization (EM) joint estimation

Surrogate modeling approximation using a mixture of experts based on Expectation-Maximization (EM) joint estimation

0.00 Avg rating0 Votes
Article ID: iaor20112398
Volume: 43
Issue: 2
Start Page Number: 243
End Page Number: 259
Publication Date: Feb 2011
Journal: Structural and Multidisciplinary Optimization
Authors: , , , ,
Keywords: design, statistics: distributions
Abstract:

An automatic method to combine several local surrogate models is presented. This method is intended to build accurate and smooth approximation of discontinuous functions that are to be used in structural optimization problems. It strongly relies on the Expectation-Maximization (EM) algorithm for Gaussian mixture models (GMM). To the end of regression, the inputs are clustered together with their output values by means of parameter estimation of the joint distribution. A local expert is then built (linear, quadratic, artificial neural network, moving least squares) on each cluster. Lastly, the local experts are combined using the Gaussian mixture model parameters found by the EM algorithm to obtain a global model. This method is tested over both mathematical test cases and an engineering optimization problem from aeronautics and is found to improve the accuracy of the approximation.

Reviews

Required fields are marked *. Your email address will not be published.