NeuroGraph - A Simulation Environment for Neural Networks, Genetic Algorithms and Fuzzy Logic
Neuro Graph is a simulation environment for neural networks. It provides an easy to use graphical user interface to design, construct and execute neural networks. The most important design goals were easy extensibility, good performance and the ability to solve real world problems. Extensibility is ensured by an extremely flexible, hierarchical internal representation (data structure), which can be easily manipulated by graphical tools. Current available extensions to the neural network simulator are components for genetic algorithms and fuzzy logic. As examples the automated generation of a neural network by a genetic algorithm and a fuzzy controller are shown. This data structure can be interpreted for debugging purposes or can be executed directly for high performance computing. In the later case C or C++ code is generated from the internal representation of the neural, gentic or fuzzy functions, compiled and stored in an executable process (library). Neuro Graph has a file or process interface to interconnect which other processes. It is also possible to generate C or C++ code representing the neural network, genetic algorithm or fuzzy component which can be incorporated in other applications.
KeywordsNeural Network Genetic Algorithm Fuzzy Logic Fuzzy System Fuzzy Controller
Unable to display preview. Download preview PDF.
- Hertz, J., Krogh, A., Palmer, R.G. (1991): Introduction to the Theory of Neural Computation, Addison-WesleyGoogle Scholar
- P. Wilke, C. Jacob,: The NeuroGraph Neural Network Simulator, Proceedings Int. Workshop on Modeling, Analysis and Simulation of Computer and Telecom-munication Systems MASCOTS93, Simulation Series Vol.25,1, San Diego, 1993, p. 341 ppGoogle Scholar
- P. Wilke, NeuroGraph V3.0 Neural Network Simulator Product Overview, WCNN’94 World Congress on Neural Networks 1994, Ed.: International Neural Network Society, 5–9 June, San Diego, Programm Addendum, p. 26–30, 1994Google Scholar
- P. Wilke, NeuroGraph V3.0 Dokumentation, Forschungsbericht des Lehrstuhls für Programmier-sprachen, Universität Erlangen-Nürnberg, 1994, ca. 200 pagesGoogle Scholar