Mathematica 9 is now available

Wolfram Library Archive

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

Recurrent Neural Networks Optimization using Continual Evolution Algorithm in Mathematica Environment

Zdeněk Buk
Organization: Czech Technical University in Prague
Department: Computational Intelligence Group
Department of Computer Science and Engineering
Faculty of Electrical Engineering

2007 Wolfram Technology Conference
Conference location

Champaign, IL


This work is focused on methods of computational intelligence, mainly recurrent neural networks and the evolutionary algorithm. Implementation of these methods in Mathematica is shown. The recurrent neural networks represent a group of artificial neural networks capable of handling time-context-based data. How to implement such time contexts in neural networks in Mathematica environment is described. The second part of the presentation describes the continual evolution algorithm as a modification of the standard genetic algorithm and its application to the recurrent neural networks optimization. Selected parts of implementation and demonstration application in Mathematica are presented.

*Applied Mathematics > Optimization
*Mathematica Technology

Download (1.4 MB) - ZIP archive