Structure optimization of neural networks for evolutionary design optimization

Authors:
M. Hüsken;Y. Jin;B. Sendhoff
Affiliations:
Ruhr-Universität Bochum, Institut für Neuroinformatik, 44780, Bochum, Germany;Honda Research Institute Europe, Institut für Neuroinformatik, Carl-Legien-Str. 30, 63073, Offenbach/Main, Germany;Honda Research Institute Europe, Institut für Neuroinformatik, Carl-Legien-Str. 30, 63073, Offenbach/Main, Germany
Venue:
Soft Computing - A Fusion of Foundations, Methodologies and Applications
Year:
2005

Citing 0
Cited 11

Generation of optimal artificial neural networks using a pattern search algorithm: Application to approximation of chemical systems

Neural Computation
ASAGA: an adaptive surrogate-assisted genetic algorithm

Proceedings of the 10th annual conference on Genetic and evolutionary computation
Evolutionary Bi-objective Learning with Lowest Complexity in Neural Networks: Empirical Comparisons

ICANNGA '07 Proceedings of the 8th international conference on Adaptive and Natural Computing Algorithms, Part I
Adaptive fuzzy fitness granulation for evolutionary optimization

International Journal of Approximate Reasoning
Meta-Modeling in Multiobjective Optimization

Multiobjective Optimization
A systems approach to evolutionary multiobjective structural optimization and beyond

IEEE Computational Intelligence Magazine
Evolutionary hidden information detection by granulation-based fitness approximation

Applied Soft Computing
Evolving robust controller parameters using covariance matrix adaptation

Proceedings of the 12th annual conference on Genetic and evolutionary computation
Performance evaluation of efficient multi-objective evolutionary algorithms for design space exploration of embedded computer systems

Applied Soft Computing
Hybrid structure based on previous knowledge and GA to search the ideal neurons quantity for the hidden layer of MLP-Application in the cold rolling process

Applied Soft Computing
Multi-objective reliability-based optimization with stochastic metamodels

Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

We study the use of neural networks as approximate models for the fitness evaluation in evolutionary design optimization. To improve the quality of the neural network models, structure optimization of these networks is performed with respect to two different criteria: One is the commonly used approximation error with respect to all available data, and the other is the ability of the networks to learn different problems of a common class of problems fast and with high accuracy. Simulation results from turbine blade optimizations using the structurally optimized neural network models are presented to show that the performance of the models can be improved significantly through structure optimization.