A parallel learning approach for neural network ensemble

Authors:
Zheng-Qun Wang;Shi-Fu Chen;Zhao-Qian Chen;Jun-Yuan Xie
Affiliations:
State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, P.R China;State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, P.R China;State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, P.R China;State Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, P.R China
Venue:
AI'04 Proceedings of the 17th Australian joint conference on Advances in Artificial Intelligence
Year:
2004

Citing 3
Cited 1

The Strength of Weak Learnability

Machine Learning
Bagging predictors

Machine Learning
Ensembling neural networks: many could be better than all

Artificial Intelligence

Integrated neural network ensemble algorithm based on clustering technology

ICONIP'06 Proceedings of the 13 international conference on Neural Information Processing - Volume Part I

Quantified Score

Hi-index	0.00

Visualization

Abstract

A component neural networks parallel training algorithm PLA is proposed, which encourages component neural network to learn from expected goal and the others, so all component neural networks are trained simultaneously and interactively In the stage of combining component neural networks, we provide a parallel weight optimal approach GASEN-e by expanding GASEN proposed by Zhou et al, which assign weight for every component neural network and bias for their ensemble Experiment results show that a neural networks ensemble system is efficient constructed by PLA and GASEN-e.