Wolfram Library Archive


Courseware Demos MathSource Technical Notes
All Collections Articles Books Conference Proceedings
Title Downloads

Hybrid Genetic Algorithm for Heterogeneous Recurrent Neural Networks Construction
Authors

Zdeněk Buk
Organization: Czech Technical University in Prague
Department: Computational Intelligence Group
Department of Computer Science and Engineering
Faculty of Electrical Engineering
URL: http://cig.felk.cvut.cz
Miroslav Snorek
Organization: Czech Technical University in Prague
Department: Department of Computer Science and Engineering
Faculty of Electrical Engineering
Conference

International Mathematica User Conference 2008
Conference location

Champaign, IL
Description

Our research is focused on optimization of recurrent neural networks using evolutionary algorithms. We are working with a hybrid evolutionary algorithm called the Continual Evolution Algorithm (CEA). Application of this algorithm to the problem of recurrent neural networks construction is the main part of our work. The CEA algorithm is used in our experiments to construct the main topology of the networks and for setting up the activation functions in particular neurons to provide complex models. The behavior of the population in the CEA and the evolution control mechanism are studied. We are presenting the methods and their implementations in our Mathematica package.
Subject

*Wolfram Technology
URL

http://www.wolfram.com/news/events/userconf2008
Downloads Download Wolfram CDF Player

Download
HybridGeneticAlgorithm_Abstract.nb (251.3 KB) - Mathematica Notebook
Download
HybridGeneticAlgorithm_Presentation.nb (5.4 MB) - Mathematica Notebook