Divide-and-conquer learning and modular perceptron networks

IEEE Transactions on Neural Networks

Volumn 12, Issue 2, 2001, Pages 250-263

  Fu, Hsin Chia ^a,b   Lee, Yen Po ^b   Chiang, Cheng Chin ^c   Pao, Hsiao Tien ^b  

^a IEEE   (Taiwan)

^b NATIONAL CHIAO TUNG UNIVERSITY   (Taiwan)

^c NATIONAL DONG HWA UNIVERSITY   (Taiwan)

Author keywords

Divide and conquer learning; Modular perceptron network; Multilayer perceptron; Weight estimation

Indexed keywords

MODULAR PERCEPTRON NETWORKS; WEIGHT ESTIMATION;

LEARNING ALGORITHMS; LEARNING SYSTEMS; MATHEMATICAL MODELS;

MULTILAYER NEURAL NETWORKS;

EID: 0035272716     PISSN: 10459227     EISSN: None     Source Type: Journal    
DOI: 10.1109/72.914522     Document Type: Article

References (29)

1. A practical gated expert network
2. Modular Neural Networks: A survey
3. A divide-and-conquer methodology for modular supervised neural network design
4. Using multithreshold quadratic sigmoidal neurons to improve classification capability of multilayer perceptrons
5. An unsupervised neural model for oriented principal component extraction
6. Boosting performance in neural networks
7. Improving performance in neural networks using boosting
8. Boosting decision trees
9. The cascade-correlation learning architecture
10. A decision-theoretic generalization of on-line learning and an application to boosting
11. Experiments with a new boosting algorithm
12. Bayesian network classifiers
13. Meiosis networks
14. Algorithms for optimal linear combinations of neural networks
15. Adaptive mixtures of local experts
16. Hierarchies mixtures of experts and the EM algorithm
17. Hybrid learning schemes for fast training of feed-forward neural networks
18. Learning to tell two spirals apart
19. Weight estimation for the learning of modular perceptron networks
20. Optimal ensemble averaging of neural network
21. Neural-network feature selector
22. The strength of weak learnability
23. A theory of the learnable
24. Node splitting: A constructive algorithm for feedforward neural networks