Parametric optimization of artificial neural networks for signal approximation applications

Authors:
J. Lane Thames;Randal Abler;Dirk Schaefer
Affiliations:
Georgia Institute of Technology, Technology Circle, Savannah, GA;Georgia Institute of Technology, Technology Circle, Savannah, GA;Georgia Institute of Technology, Technology Circle, Savannah, GA
Venue:
Proceedings of the 49th Annual Southeast Regional Conference
Year:
2011

Citing 4
Cited 0

Genetic Algorithms in Search, Optimization and Machine Learning

Genetic Algorithms in Search, Optimization and Machine Learning
Machine Learning

Machine Learning
Evolving neural networks through augmenting topologies

Evolutionary Computation
Bio-Inspired Parameter Tunning of MLP Networks for Gene Expression Analysis

HIS '08 Proceedings of the 2008 8th International Conference on Hybrid Intelligent Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Artificial neural networks are used to solve diverse sets of problems. However, the accuracy of the network's output for a given problem domain depends on appropriate selection of training data as well as various design parameters that define the structure of the network before it is trained. Genetic algorithms have been used successfully for many types of optimization problems. In this paper, we describe a methodology that uses genetic algorithms to find an optimal set of configuration parameters for artificial neural networks such that the network's approximation error for signal approximation problems is minimized.