A modified genetic algorithm for fast training neural networks

Authors:
Dongsun Kim;Hyunsik Kim;Duckjin Chung
Affiliations:
DMB Project Office, Korea Electronics Technology Institute, PyungTaeksi, KyungGido, Korea;DMB Project Office, Korea Electronics Technology Institute, PyungTaeksi, KyungGido, Korea;Information Technology and Telecommunications, INHA University, Incheon, Korea
Venue:
ISNN'05 Proceedings of the Second international conference on Advances in Neural Networks - Volume Part I
Year:
2005

Citing 2
Cited 3

Optimizing neural networks using faster, more accurate genetic search

Proceedings of the third international conference on Genetic algorithms
Learning in linear neural networks: a survey

IEEE Transactions on Neural Networks

Investigation of bagging ensembles of genetic neural networks and fuzzy systems for real estate appraisal

ACIIDS'11 Proceedings of the Third international conference on Intelligent information and database systems - Volume Part II
Empirical comparison of resampling methods using genetic neural networks for a regression problem

HAIS'11 Proceedings of the 6th international conference on Hybrid artificial intelligent systems - Volume Part II
Training neural networks with harmony search algorithms for classification problems

Engineering Applications of Artificial Intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

The training of feed-forward Neural Networks (NNs) by backpropagation (BP) is much time-consuming and complex task of great importance. To overcome this problem, we apply Genetic Algorithm (GA) to determine parameters of NN automatically and propose a efficient GA which reduces its iterative computation time for enhancing the training capacity of NN. Proposed GA is based on steady-state model among continuous generation model and used the modified tournament selection, as well as special survival condition. To show the validity of the proposed method, we compare with conventional and the survival-based GA using mathematical optimization problems and set covering problem. In addition, we estimate the performance of training the layered feedforward NN with GA and BP.