Modified Levenberg-Marquardt Method For Neural Networks Training

Abstract

In this paper a modification on Levenberg-Marquardt algorithm for MLP neural network learning is proposed. The proposed algorithm has good convergence. This method reduces the amount of oscillation in learning procedure. An example is given to show usefulness of this method. Finally a simulation verifies the results of proposed method.

References

Page 1

	Year	Citations
Learning representations by back-propagating errors David E. Rumelhart, Geoffrey E. Hinton, Ronald J. Williams Nature EngineeringMachine LearningFeature LearningPattern RecognitionKnowledge Discovery	1986	29.7K
Increased rates of convergence through learning rate adaptation Robert A. Jacobs Neural Networks Rate AdaptationComputational Learning TheoryMachine LearningConvergence AnalysisAdaptive Optimization	1988	1.8K
First- and Second-Order Methods for Learning: Between Steepest Descent and Newton's Method Roberto Battiti Neural Computation Mathematical ProgrammingNumerical AnalysisEngineeringMachine LearningAlgorithmic Learning	1992	1.2K
Conjugate gradient algorithm for efficient training of artificial neural networks Charalambos A. Charalambous IEE Proceedings G Circuits Devices and Systems Artificial IntelligenceModel OptimizationConjugate Gradient AlgorithmEvolving Neural NetworkEngineering	1992	374
Accelerated learning in layered neural networks Sara A. Solla, Esther Levin, Michael Fleisher Complex Systems Model OptimizationAbst RactMini MizEngineeringMachine Learning	1988	211
Efficient algorithm for training neural networks with one hidden layer Bogdan M. Wilamowski, Yixin Chen, A. Malinowski Artificial IntelligenceNumerical AnalysisEngineeringMachine LearningFeedforward Neural Networks	2003	89
Acceleration of Levenberg-Marquardt training of neural networks with variable decay rate Taicong Chen, Dajian Han, F.T.K. Au, Numerical AnalysisEngineeringMachine LearningComputational Learning TheoryComputational Neuroscience	2004	71
MEKA-a fast, local algorithm for training feedforward neural networks Sahil Shah, F. Palmieri Nonlinear System IdentificationEvolving Neural NetworkEngineeringMachine LearningConventional Backpropagation Algorithm	1990	64

Page 1