A system and method for generating a neural network ensemble. Conventionalalgorithms are used to train a number of neural networks having errordiversity, for example by having a different number of hidden nodes in eachnetwork. A genetic algorithm having a multi-objective fitness function is usedto select one or more ensembles. The fitness function includes a negativeerror correlation objective to insure diversity among the ensemble members. Agenetic algorithm may be used to select weighting factors for the multi-objective function. In one application, a trained model may be used to producesynthetic open hole logs in response to inputs of cased hole log data.
展开▼
