On the optimal structure design of multilayer feedforward neural networks for pattern recognition

International Journal of Pattern Recognition and Artificial Intelligence

Volumn 16, Issue 4, 2002, Pages 375-398

  Chen, Daqing ^a   Burrell, Phillip ^a  

^a LONDON SOUTH BANK UNIVERSITY   (United Kingdom)

Author keywords

Constructive algorithm; Discriminant decision; Genetic algorithm; Internal representation; Machine learning; Multilayer feedforward neural network; Multilayer perceptron; Pruning algorithm; Support vector machine

Indexed keywords

DECISION THEORY; EIGENVALUES AND EIGENFUNCTIONS; FEATURE EXTRACTION; GENETIC ALGORITHMS; LEARNING SYSTEMS; MATRIX ALGEBRA; PATTERN RECOGNITION;

CONSTRUCTIVE ALGORITHM; DISCRIMINANT DECISION; INTERNAL REPRESENTATION; MULTILAYER FEEDFORWARD NEURAL NETWORKS; PRUNING ALGORITHM; SUPPORT VECTOR MACHINE;

MULTILAYER NEURAL NETWORKS;

EID: 0036601225     PISSN: 02180014     EISSN: None     Source Type: Journal    
DOI: 10.1142/S0218001402001812     Document Type: Article

References (54)

1. Bounds on the number of hidden units in binary-valued three-layer neural networks
2. Recursive dynamic node creation in multilayer neural networks
3. Enhanced training algorithms, and integrated training/architecture selection for multilayer perceptron networks
4. Comparison of view-based object recognition algorithm using realistic 3d models
5. Support vector machines for histogram-based image classification
6. Study and application of the nonlinear mechanism of the multilayer feedforward networks
7. Theory of multilayer feedforward neural networks with finite nodes
8. Support-vector networks
9. Approximation by superposition of a sigmoidal function
10. On the approximate realization of continuous mappings by neural networks
11. Multilayer perceptrons and data analysis
12. On the relations between discriminant analysis and multilayer perceptron
13. The differentiation of pseudo-inverses and nonlinear least squares problems whose variables separate
14. Intelligent hybrid systems: Issues, classifications and future directions
15. Classification trees with neural network feature extraction
16. Neural classifier construction using regularization, pruning and test error estimation
17. Back-propagation algorithm which varies the number of hidden units
18. Multilayer feedforward networks are universal approximation
19. Approximation capabilities of multilayer feedforward networks
20. Some new results on neural network approximation
21. Upper bounds on the number of hidden neurons in feedforward networks with arbitrary bounded nonlinear activation functions
22. Bounds on the number of hidden neurons in multilayer perceptrons
23. Genetic algorithms and their applications to the design of neural networks
24. Constructive algorithms for structure learning in feedforward neural networks for regression problems
25. Feature extraction based on decision boundaries
26. Decision boundary feature extraction for neural networks
27. Training MLPs layer by layer using an objective function for internal representations
28. An introduction to computing with neural nets
29. Optimized feature extraction and the Bayes decision in feed-forward classifier networks
30. Genetic evolution of the topology and weight distribution of neural networks
31. Bounds on the number of samples needed for neural learning
32. On hidden nodes for neural nets
33. Genetic set recombination and its application to neural network topology optimisation
34. Pruning algorithms - A survey
35. Divide and conquer neural networks
36. Learning internal representation by error propagation
37. Evolutionary algorithms and gradient search: Similarities and differences
38. A neural network construction algorithm with application to image compression
39. Linear independence of internal representations in multilayer perceptrons
40. Separability of internal representations in multilayer perceptrons with application to learning
41. Capabilities of a four-layered feedforward neural network: Four layer versus three
42. An overview of statistical learning theory
43. The optimal internal representation of multilayer classifier networks performs nonlinear discriminant analysis
44. Multiclass support vector machines
45. Node splitting: A constructive algorithm for feed-forward neural networks
46. A new evolutionary system for evolving artificial neural networks
47. Research on the theory and methods of multilayer feedforward neural networks for pattern recognition
48. The construction of minimal multilayered perceptrons: A case study for sorting