Using Model Trees for Classification
Machine Learning
MultiBoosting: A Technique for Combining Boosting and Wagging
Machine Learning
The constraint based decomposition (CBD) training architecture
Neural Networks
Using Correspondence Analysis to Combine Classifiers
Machine Learning
Generating Accurate Rule Sets Without Global Optimization
ICML '98 Proceedings of the Fifteenth International Conference on Machine Learning
Neural Networks - 2003 Special issue: Advances in neural networks research IJCNN'03
Cooperative Coevolution: An Architecture for Evolving Coadapted Subcomponents
Evolutionary Computation
Evolutionary ensembles with negative correlation learning
IEEE Transactions on Evolutionary Computation
Inducing oblique decision trees with evolutionary algorithms
IEEE Transactions on Evolutionary Computation
The balance between proximity and diversity in multiobjective evolutionary algorithms
IEEE Transactions on Evolutionary Computation
Cooperative coevolution of artificial neural network ensembles for pattern classification
IEEE Transactions on Evolutionary Computation
Designing RBFNNs using prototype selection
MCPR'10 Proceedings of the 2nd Mexican conference on Pattern recognition: Advances in pattern recognition
Hi-index | 0.00 |
This article presents a new learning algorithm, CO-RBFNN, for complex classifications, which attempts to construct the radial basis function neural network (RBFNN) models by using a cooperative coevolutionary algorithm (Co-CEA). The Co-CEA utilizes a divide-and-cooperative mechanism by which subpopulations are coevolved in separate populations of evolutionary algorithms executing in parallel. A modified K-means method is employed to divide the initial hidden nodes into modules that are represented as subpopulation of the Co-CEA. Collaborations among the modules are formed to obtain complete solutions. The algorithm adopts a matrix-form mixed encoding to represent the RBFNN hidden layer structure, the optimum of which is achieved by coevolving all parameters. Experimental results on eight UCI datasets illustrate that CO-RBFNN is able to produce a higher accuracy of classification with a much simpler network structure in fewer evolutionary trials when compared with other alternative standard algorithms.