GENNET-Toolbox: an evolving genetic algorithm for neural network training

Authors:
Vicente Gómez-Garay;Eloy Irigoyen;Fernando Artaza
Affiliations:
Automatic Control and System Engineering Department, University of the Basque Country, E.T.S.I., Bilbao, Spain;Automatic Control and System Engineering Department, University of the Basque Country, E.T.S.I., Bilbao, Spain;Automatic Control and System Engineering Department, University of the Basque Country, E.T.S.I., Bilbao, Spain
Venue:
HAIS'10 Proceedings of the 5th international conference on Hybrid Artificial Intelligence Systems - Volume Part I
Year:
2010

Citing 5
Cited 0

Adaptation in natural and artificial systems

Adaptation in natural and artificial systems
Genetic algorithms + data structures = evolution programs (3rd ed.)

Genetic algorithms + data structures = evolution programs (3rd ed.)
Toward global optimization of neural networks: a comparison of the genetic algorithm and backpropagation

Decision Support Systems
Genetic Algorithms in Search, Optimization and Machine Learning

Genetic Algorithms in Search, Optimization and Machine Learning
Introduction to Evolutionary Computing

Introduction to Evolutionary Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Genetic Algorithms have been used from 1989 for both Neural Network training and design Nevertheless, the use of a Genetic Algorithm for adjusting the Neural Network parameters can still be engaging This work presents the study and validation of a different approach to this matter by introducing a Genetic Algorithm designed for Neural Network training This algorithm features a mutation operator capable of working on three levels (network, neuron and layer) and with the mutation parameters encoded and evolving within each individual We also explore the use of three types of hybridization: post-training, Lamarckian and Baldwinian These proposes in combination with the algorithm, show for a fast and powerful tool for Neural Network training.