Title :
Yet another genetic algorithm for feed-forward neural networks
Author_Institution :
Inst. of Comput. Sci., Czechoslovak Acad. of Sci., Prague, Czech Republic
Abstract :
A functional equivalence property of feedforward networks has been proposed to reduce the search space of learning algorithms. We summarize previous results, describing the form of functional equivalence for one-hidden-layer perceptron networks and radial basis function (RBF) networks with Gaussians. The description of equivalence classes is used in a proposition of a genetic learning algorithm which is tested on two simple problems and which outperforms the standard genetic learning procedure
Keywords :
equivalence classes; feedforward neural nets; genetic algorithms; learning (artificial intelligence); perceptrons; search problems; software performance evaluation; Gaussians; equivalence classes; feedforward neural networks; functional equivalence property; genetic algorithm; genetic learning procedure; learning algorithm; perceptrons; performance; radial basis function networks; search space reduction; Computer networks; Computer science; Feedforward neural networks; Feedforward systems; Gaussian processes; Genetic algorithms; Neural networks; Radial basis function networks; Testing;
Conference_Titel :
Tools with Artificial Intelligence, 1997. Proceedings., Ninth IEEE International Conference on
Conference_Location :
Newport Beach, CA
Print_ISBN :
0-8186-8203-5
DOI :
10.1109/TAI.1997.632278
