A comparison of different methods for combining multiple neural networks models

Proceedings of the International Joint Conference on Neural Networks

Volumn 1, Issue , 2002, Pages 828-833

  Ahmad, Zainal ^a   Zhang, Jie ^a  

^a NEWCASTLE UNIVERSITY   (United Kingdom)

Author keywords

[No Author keywords available]

Indexed keywords

COMPUTER SIMULATION; DATA STRUCTURES; NONLINEAR CONTROL SYSTEMS; REGRESSION ANALYSIS; ROBUSTNESS (CONTROL SYSTEMS);

BOOTSTRAP AGGREGATED NETWORKS; COMBINING MULTIPLE NEURAL NETWORKS;

NEURAL NETWORKS;

EID: 0036073711     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: None     Document Type: Conference Paper

References (22)

1. Overfitting in neural networks: Backpropagation, conjugate gradient and early stopping
2. Regularisation learning and early stopping in linear networks
3. Optimal linear combination
4. Treating harmful collinearity in neural networks ensembles
5. Adaptive mixture of local expert
6. Hierarchical mixtures of expert and the EM algorithm
7. Dynamics of pH in controlled stirred tank reactor
8. Improving neural networks training solution using regularisation
9. Generalisation and parameter estimation in feedforward nets: Some experiments
10. Robust control for non-linear system by universal learning networks considering fuzzy criterion and second order derivatives
11. When networks disagree: Ensembles methods for hybrid neural networks
12. Multi nets system
13. An information theoretic approach for combining neural network process models
14. Process modelling using stacked neural networks
15. Stacked generalisation
16. Developing robust neural network models by using both dynamic and static process operating data
17. Inferential estimation of polymer quality using bootstrap aggregated neural networks
18. Developing robust non-linear models through bootstrap aggregated neural networks
19. Inferential estimation of polymer quality using stacked neural networks
20. Prediction of polymer quality in batch polymerisation reactors using robust neural networks